{ "cells": [ { "cell_type": "code", "execution_count": 1, "metadata": { "slideshow": { "slide_type": "skip" } }, "outputs": [], "source": [ "%%capture\n", "%load_ext autoreload\n", "%autoreload 2\n", "import sys\n", "sys.path.append(\"../statnlpbook/\")\n", "#import util\n", "import ie\n", "import tfutil\n", "import random\n", "import numpy as np\n", "import tensorflow as tf\n", "np.random.seed(1337)\n", "tf.set_random_seed(1337)\n", "\n", "#util.execute_notebook('relation_extraction.ipynb')" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "skip" } }, "source": [ "\n", "$$\n", "\\newcommand{\\Xs}{\\mathcal{X}}\n", "\\newcommand{\\Ys}{\\mathcal{Y}}\n", "\\newcommand{\\y}{\\mathbf{y}}\n", "\\newcommand{\\balpha}{\\boldsymbol{\\alpha}}\n", "\\newcommand{\\bbeta}{\\boldsymbol{\\beta}}\n", "\\newcommand{\\aligns}{\\mathbf{a}}\n", "\\newcommand{\\align}{a}\n", "\\newcommand{\\source}{\\mathbf{s}}\n", "\\newcommand{\\target}{\\mathbf{t}}\n", "\\newcommand{\\ssource}{s}\n", "\\newcommand{\\starget}{t}\n", "\\newcommand{\\repr}{\\mathbf{f}}\n", "\\newcommand{\\repry}{\\mathbf{g}}\n", "\\newcommand{\\x}{\\mathbf{x}}\n", "\\newcommand{\\prob}{p}\n", "\\newcommand{\\a}{\\alpha}\n", "\\newcommand{\\b}{\\beta}\n", "\\newcommand{\\vocab}{V}\n", "\\newcommand{\\params}{\\boldsymbol{\\theta}}\n", "\\newcommand{\\param}{\\theta}\n", "\\DeclareMathOperator{\\perplexity}{PP}\n", "\\DeclareMathOperator{\\argmax}{argmax}\n", "\\DeclareMathOperator{\\argmin}{argmin}\n", "\\newcommand{\\train}{\\mathcal{D}}\n", "\\newcommand{\\counts}[2]{\\#_{#1}(#2) }\n", "\\newcommand{\\length}[1]{\\text{length}(#1) }\n", "\\newcommand{\\indi}{\\mathbb{I}}\n", "$$" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "# Relation Extraction" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "## Motivation \n", "\n", "* The amount of available information is growing exponentially\n", "* Text contains a lot of information\n", "* Only some of information is relevant for each use case\n", "* How can we automatically make sense of information?" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "**Information Extraction** addresses this\n", "\n", "[Alchemy information extraction demo](https://alchemy-language-demo.mybluemix.net/)\n", "\n", "[ReVerb demo](http://openie.allenai.org/)" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "## Subtasks of Information Extraction\n", "\n", "* **Document** Classification:\n", " * Assign a label to each document, often representing the topic\n", " * Often treated as standard [classification task](chapters/doc_classify.ipynb)\n", "* **Named Entity Recognition (NER)**:\n", " * Recognise boundaries of entities in text, e.g. \"New York\", \"New York Times\" \n", "* **Named Entity Classification (NEC)**:\n", " * Assign a type to each entity (e.g. \"New York\" -> LOC, \"New York Times\" -> ORG)" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "## Named Entity Recognition and Classification\n", " \n", "* NER and NEC are often approached together (NERC)\n", "* They are often treated as a [sequence labelling task](chapters/sequence_labelling.ipynb)\n", "* Every token is assigned \n", " * a training label (e.g. PER)\n", " * and sometimes a positional tag (e.g. begininning of sequence (B), inside sequence (I), outside sequence (O)\n", "* At test time, subsequent tokens with the same label are merged to one sequence" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "## Named Entity Recognition and Classification: Example\n", " \n", "\n", "| Sebastian | Riedel | is | a | reader | at | University | College | London | \n", "|-|-|-|-|-|-|-|-|-|\n", "| B-PER | I-PER | O | O | O | O | B-LOC | I-LOC | I-LOC | \n", "\n", " Sebastian Riedel: PER \n", " University College London: LOC " ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "## Subtasks of Information Extraction\n", "\n", "* **Relation** Extraction:\n", " * Recognise relatios between entities, e.g. \"S. Riedel reader-at UCL\"\n", " * classification task\n", "* **Temporal** Information Extraction:\n", " * Recognise and/or normalise temporal expressions, e.g. \"tomorrow morning at 8\" -> \"2016-11-26 08:00:00\"\n", " * sequence labelling task" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "## Subtasks of Information Extraction\n", "\n", "* **Event** Extraction:\n", " * Recognise events, typically consisting of entities and relations between them at a point in time and place, e.g. an election\n", " * classification task" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "### Relation Extraction\n", "\n", "Task of extracting **semantic relations between arguments**\n", "* Arguments are entities\n", " * general concepts such as \"a company\" (ORG), \"a person\" (PER), \"a location\" (LOC)\n", " * instances of such concepts (e.g. \"Microsoft\", \"Bill Gates\"), which are called proper names or named entitites (NEs)\n", "* Relation extraction builds on the task of named entity recognition" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "Relation extraction is relevant for many high-level NLP tasks, such as\n", "* for question answering, where users ask questions such as \"Who founded Microsoft?\",\n", "* for information retrieval, which often relies on large collections of structured information as background data, and\n", "* for text and data mining, where larger patterns in relations between concepts are discovered, e.g. temporal patterns about startups" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "## Relation Extraction as Structured Prediction\n", "We can formalise relation extraction as an instance of [structured prediction](chapters/structured_prediction.ipynb)\n", "* The input space $\\mathcal{X}$ are pairs of arguments $\\mathcal{E}$ and supporting texts $\\mathcal{S}$ those arguments appear in\n", "* The output space $\\mathcal{Y}$ is a set of relation labels such as $\\Ys=\\{ \\text{founder-of},\\text{employee-at},\\text{professor-at},\\text{NONE}\\}$. " ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "## Relation Extraction as Structured Prediction\n", "* The goal is to define a model \\\\(s_{\\params}(\\x,y)\\\\) that assigns high *scores* to the label $\\mathcal{y}$ that fits the arguments and supporting text $\\mathcal{x}$, and lower scores otherwise. \n", "* The model will be parametrized by \\\\(\\params\\\\), and these parameters we will learn from some training set of $\\mathcal{x,y}$ pairs\n", "* When we need to classify input instances $\\mathcal{x}$ consisting again of pairs of arguments and supporting texts, we have to solve the maximisation problem $\\argmax_y s_{\\params}(\\x,y)$." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "## Relation Extraction Approaches\n", "* **Pattern-Based** Relation Extraction:\n", " * Extract relations via manually defined textual pattern matching\n", "* **Bootstrapping**:\n", " * Learn to extract relations via manually defined textual patterns, and use those to find more patterns and so forth, iteratively" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "## Relation Extraction Approaches\n", "* **Supervised** Relation Extraction:\n", " * Train a supervised model, from manually labelled training examples, to extract relations\n", "* **Distantly Supervised** Relation Extraction:\n", " * Automatically annotate training data for supervised relation extraction, based on entries in a knowledge base" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "## Relation Extraction Approaches\n", "* **Universal Schema** Relation Extraction:\n", " * Model relation types and their surface forms in the same space, possible method for combining pattern-based, supervised and distantly supervised relation extraction" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "\n", "## Relation Extraction Example\n", "* Extracting \"method used for task\" relations from sentences in computer science publications\n", "* The first step would normally be to detection named entities, i.e. to determine tose pairs of arguments $\\mathcal{E}$. For simplicity, our training data already contains those annotations.\n" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "## Pattern-Based Extraction\n", "* The simplest relation extraction model defines a set of textual patterns for each relation and then assigns labels to entity pairs whose sentences match that pattern. \n", "* The training data consists of entity pairs $\\mathcal{E}$, patterns $A$ and labels $Y$." ] }, { "cell_type": "code", "execution_count": 2, "metadata": { "slideshow": { "slide_type": "subslide" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Training patterns and entity pairs for relation 'method used for task'\n" ] }, { "data": { "text/plain": [ "[('demonstrates XXXXX and clustering techniques for XXXXX',\n", " ['text mining', 'building domain ontology']),\n", " ('demonstrates text mining and XXXXX for building XXXXX',\n", " ['clustering techniques', 'domain ontology']),\n", " ('the XXXXX is able to enhance the XXXXX',\n", " ['ensemble classifier', 'detection of construction materials'])]" ] }, "execution_count": 2, "metadata": {}, "output_type": "execute_result" } ], "source": [ "training_patterns, training_entpairs = ie.readLabelledPatternData()\n", "print(\"Training patterns and entity pairs for relation 'method used for task'\")\n", "[(tr_a, tr_e) for (tr_a, tr_e) in zip(training_patterns[:3], training_entpairs[:3])]" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "* The patterns are sentences where the entity pairs where blanked with the placeholder 'XXXXX'\n", "* For the training data, we also have labels\n", "* There are only one positive label, 'method used for task'\n", "* For the testing data, we do not know the relations for the instances" ] }, { "cell_type": "code", "execution_count": 3, "metadata": { "slideshow": { "slide_type": "subslide" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Testing patterns and entity pairs\n" ] }, { "data": { "text/plain": [ "[('a method for estimation of XXXXX of XXXXX is presented',\n", " ['effective properties', 'porous materials']),\n", " ('accounting for XXXXX is essential for estimation of XXXXX',\n", " ['nonlinear effects', 'effective properties']),\n", " ('develops the heterogeneous XXXXX for fiber-reinforced XXXXX',\n", " ['feature model', 'object modeling'])]" ] }, "execution_count": 3, "metadata": {}, "output_type": "execute_result" } ], "source": [ "testing_patterns, testing_entpairs = ie.readPatternData()\n", "print(\"Testing patterns and entity pairs\")\n", "[(tr_a, tr_e) for (tr_a, tr_e) in zip(testing_patterns[0:3], testing_entpairs[:3])]" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "* We build a scoring model to determine which of the testing instances are examples for the relation 'method used for task' and which ones are not\n", " * A pattern scoring model \\\\(s_{\\params}(\\x,y)\\\\) only has one parameter\n", " * It assignes scores to each relation label \\\\(y\\\\) proportional to the matches with the set of textual patterns\n", " * The final label assigned to each instance is then the one with the highest score." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "* Here, our pattern scoring model is even simpler since we only have patterns for one relation\n", " * The final label assigned to each instance is 'method used for task' if there is a match with a pattern, and 'NONE' if there is no match." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "Let's have a closer look at how pattern matching works:\n", "* Recall that the original patterns in the training data are sentences where the entity pairs are blanked with 'XXXXX'\n", "* We could use those patterns to find new sentences\n", "* However, we are not likely to find many since the patterns are very specific to the example\n", "* We need to generalise those patterns to less specific ones\n", "* A simple way is to define the sequence of words between each entity pair as a pattern, like so:" ] }, { "cell_type": "code", "execution_count": 4, "metadata": { "slideshow": { "slide_type": "subslide" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "demonstrates XXXXX and clustering techniques for XXXXX\n" ] }, { "data": { "text/plain": [ "'and clustering techniques for'" ] }, "execution_count": 4, "metadata": {}, "output_type": "execute_result" } ], "source": [ "def sentenceToShortPath(sent):\n", " \"\"\"\n", " Returns the path between two arguments in a sentence, where the arguments have been masked\n", " Args:\n", " sent: the sentence\n", " Returns:\n", " the path between to arguments\n", " \"\"\"\n", " sent_toks = sent.split(\" \")\n", " indeces = [i for i, ltr in enumerate(sent_toks) if ltr == \"XXXXX\"]\n", " pattern = \" \".join(sent_toks[indeces[0]+1:indeces[1]])\n", " return pattern\n", "\n", "print(training_patterns[0])\n", "sentenceToShortPath(training_patterns[0])" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "* There are many different alternatives to this method shortening patterns.\n", "* **Thought exercise**: \n", " * what is a possible problem with this way of shortening patterns and what are better ways of generalising patterns?" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "* After the sentences shortening / pattern generalisation is defined, we can then apply those patterns to testing instances to classify them into 'method used for task' and 'NONE'\n", "* In the example here, we return the instances which contain a 'method used for task' pattern" ] }, { "cell_type": "code", "execution_count": 5, "metadata": { "slideshow": { "slide_type": "subslide" } }, "outputs": [ { "data": { "text/plain": [ "['paper reviews applications of XXXXX in XXXXX',\n", " 'a novel approach was developed to determine the XXXXX in XXXXX',\n", " 'four different types of insoles were examined in terms of their effects on XXXXX in XXXXX',\n", " 'the findings can aid in better understanding the insole design features that could improve XXXXX in XXXXX',\n", " 'this new approach provides more degrees of freedom and XXXXX in XXXXX']" ] }, "execution_count": 5, "metadata": {}, "output_type": "execute_result" } ], "source": [ "def patternExtraction(training_sentences, testing_sentences):\n", " \"\"\"\n", " Given a set of patterns for a relation, searches for those patterns in other sentences\n", " Args:\n", " sent: training sentences with arguments masked, testing sentences with arguments masked\n", " Returns:\n", " the testing sentences which the training patterns appeared in\n", " \"\"\"\n", " # convert training and testing sentences to short paths to obtain patterns\n", " training_patterns = set([sentenceToShortPath(test_sent) for test_sent in training_sentences])\n", " testing_patterns = [sentenceToShortPath(test_sent) for test_sent in testing_sentences]\n", " # look for training patterns in testing patterns\n", " testing_extractions = []\n", " for i, testing_pattern in enumerate(testing_patterns):\n", " if testing_pattern in training_patterns: # look for exact matches of patterns\n", " testing_extractions.append(testing_sentences[i])\n", " return testing_extractions\n", "\n", "patternExtraction(training_patterns[:300], testing_patterns[:300])" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "* One of the shortcomings of this pattern-based approach is that the set of patterns has to be defined manually\n", "* Also, the model does not learn new patterns\n", "* We will next look at an approach which addresses those two shortcomings\n" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "## Bootstrapping\n", "\n", "* Bootstrapping relation extraction models take the same input as pattern-based approaches\n", " * a set of entity pairs and patterns\n", "* Overall idea: extract more patterns and entity pairs iteratively\n", "* For this, we need two helper methods: \n", " * one that generalises from patterns to extract more patterns and entity pairs\n", " * another one that generalises from entity pairs to extract more patterns and entity pairs\n" ] }, { "cell_type": "code", "execution_count": 6, "metadata": { "slideshow": { "slide_type": "subslide" } }, "outputs": [], "source": [ "# generalises from patterns to extract more patterns and entity pairs\n", "def searchForPatternsAndEntpairsByPatterns(training_patterns, testing_patterns, testing_entpairs, testing_sentences):\n", " testing_extractions = []\n", " appearing_testing_patterns = []\n", " appearing_testing_entpairs = []\n", " for i, testing_pattern in enumerate(testing_patterns):\n", " if testing_pattern in training_patterns: # if there is an exact match of a pattern\n", " testing_extractions.append(testing_sentences[i])\n", " appearing_testing_patterns.append(testing_pattern)\n", " appearing_testing_entpairs.append(testing_entpairs[i])\n", " return testing_extractions, appearing_testing_patterns, appearing_testing_entpairs" ] }, { "cell_type": "code", "execution_count": 7, "metadata": { "slideshow": { "slide_type": "subslide" } }, "outputs": [], "source": [ "# generalises from entity pairs to extract more patterns and entity pairs\n", "def searchForPatternsAndEntpairsByEntpairs(training_entpairs, testing_patterns, testing_entpairs, testing_sentences):\n", " testing_extractions = []\n", " appearing_testing_patterns = []\n", " appearing_testing_entpairs = []\n", " for i, testing_entpair in enumerate(testing_entpairs):\n", " if testing_entpair in training_entpairs: # if there is an exact match of an entity pair\n", " testing_extractions.append(testing_sentences[i])\n", " appearing_testing_entpairs.append(testing_entpair)\n", " appearing_testing_patterns.append(testing_patterns[i])\n", " return testing_extractions, appearing_testing_patterns, appearing_testing_entpairs" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "The two helper functions are then applied iteratively:" ] }, { "cell_type": "code", "execution_count": 8, "metadata": { "slideshow": { "slide_type": "subslide" } }, "outputs": [], "source": [ "def bootstrappingExtraction(train_sents, train_entpairs, test_sents, test_entpairs, num_iter=10):\n", " \"\"\"\n", " Given a set of patterns and entity pairs for a relation, extracts more patterns and entity pairs iteratively\n", " Args:\n", " train_sents: training sentences with arguments masked\n", " train_entpairs: training entity pairs\n", " test_sents: testing sentences with arguments masked\n", " test_entpairs: testing entity pairs\n", " Returns:\n", " the testing sentences which the training patterns or any of the inferred patterns appeared in\n", " \"\"\"\n", "\n" ] }, { "cell_type": "code", "execution_count": 9, "metadata": { "slideshow": { "slide_type": "subslide" } }, "outputs": [], "source": [ "def bootstrappingExtraction(train_sents, train_entpairs, test_sents, test_entpairs, num_iter=10):\n", " # convert training and testing sentences to short paths to obtain patterns\n", " train_patterns = set([sentenceToShortPath(s) for s in train_sents])\n", " train_patterns.discard(\"in\") # too general, remove this\n", " test_patterns = [sentenceToShortPath(s) for s in test_sents]\n", "\n", " # iteratively get more patterns and entity pairs\n", " for i in range(1, num_iter):\n", " print(\"Number extractions at iteration\", str(i), \":\", str(len(test_extracts)))\n", " print(\"Number patterns at iteration\", str(i), \":\", str(len(train_patterns)))\n", " print(\"Number entpairs at iteration\", str(i), \":\", str(len(train_entpairs)))\n", " # get more patterns and entity pairs\n", " test_extracts_p, ext_test_patterns_p, ext_test_entpairs_p = searchForPatternsAndEntpairsByPatterns(train_patterns, test_patterns, test_entpairs, test_sents)\n", " test_extracts_e, ext_test_patterns_e, ext_test_entpairs_e = searchForPatternsAndEntpairsByEntpairs(train_entpairs, test_patterns, test_entpairs, test_sents)\n", " # add them to the existing entity pairs for the next iteration\n", " train_patterns.update(ext_test_patterns_p)\n", " train_patterns.update(ext_test_patterns_e)\n", " train_entpairs.extend(ext_test_entpairs_p)\n", " train_entpairs.extend(ext_test_entpairs_e)\n", " test_extracts.extend(test_extracts_p)\n", " test_extracts.extend(test_extracts_e)\n", "\n", " return test_extracts, test_entpairs" ] }, { "cell_type": "code", "execution_count": 10, "metadata": { "slideshow": { "slide_type": "subslide" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Number extractions at iteration 0 : 0\n", "Number patterns at iteration 0 : 19\n", "Number entpairs at iteration 0 : 22\n", "Number extractions at iteration 1 : 2\n", "Number patterns at iteration 1 : 19\n", "Number entpairs at iteration 1 : 24\n", "Number extractions at iteration 2 : 6\n", "Number patterns at iteration 2 : 19\n", "Number entpairs at iteration 2 : 28\n", "Number extractions at iteration 3 : 10\n", "Number patterns at iteration 3 : 19\n", "Number entpairs at iteration 3 : 32\n", "Number extractions at iteration 4 : 14\n", "Number patterns at iteration 4 : 19\n", "Number entpairs at iteration 4 : 36\n", "Number extractions at iteration 5 : 18\n", "Number patterns at iteration 5 : 19\n", "Number entpairs at iteration 5 : 40\n" ] } ], "source": [ "test_extracts, test_entpairs = ie.bootstrappingExtraction(training_patterns, training_entpairs, testing_patterns, testing_entpairs)" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "One of the things noticable is that with each iteration, the number of extractions we find increases, but they are less correct." ] }, { "cell_type": "code", "execution_count": 11, "metadata": { "slideshow": { "slide_type": "subslide" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "a strongly XXXXX is proposed to solve the XXXXX ['nonlinear effects', 'effective properties']\n", "the novelties of our work are in both theory and application . we propose a new distance formula for interval type-2 fuzzy sets . based on the proposed distance formula , we propose a new XXXXX to solve a XXXXX and finally to illustrate the applicability of the proposed method , a case study is used ['feature model', 'object modeling']\n", "\n", "a strongly XXXXX is proposed to solve the XXXXX ['sample size', 'gps data']\n", "the novelties of our work are in both theory and application . we propose a new distance formula for interval type-2 fuzzy sets . based on the proposed distance formula , we propose a new XXXXX to solve a XXXXX and finally to illustrate the applicability of the proposed method , a case study is used ['memetic algorithm', 'genetic algorithm']\n" ] } ], "source": [ "for (s, e) in zip(test_extracts[1:3], test_entpairs[1:3]):\n", " print(s, e)\n", "print(\"\")\n", "for (s, e) in zip(test_extracts[-3:-1], test_entpairs[-3:-1]):\n", " print(s, e)" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "* One of the reasons is that the semantics of the pattern shifts\n", " * here we try to find new patterns for 'method used for task'\n", " * but because the instances share a similar context with other relations, the patterns and entity pairs iteratively move away from the 'method used in task' relation\n", " * Another example: 'student-at' and 'lecturere-at' relations, that have many overlapping contexts\n" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "* One way of improving this is with confidence values for each entity pair and pattern\n", " * For example, we might want to avoid entity pairs or patterns which are too general and penalise them" ] }, { "cell_type": "code", "execution_count": 12, "metadata": { "slideshow": { "slide_type": "subslide" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Counter({'to solve a': 11, 'is proposed to solve the': 11})\n" ] } ], "source": [ "from collections import Counter\n", "te_cnt = Counter()\n", "for te in test_extracts:\n", " te_cnt[sentenceToShortPath(te)] += 1\n", "print(te_cnt)" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "* Such as 'noisy' pattern is e.g. the 'in' pattern was found originally, which maches many contexts that are not 'method used for task' \n", "* **Thought exercise**: \n", " * how would a confidence weighting for patterns work here?" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "## Supervised Relation Extraction\n", "* A different way of assigning a relation label to new instances is to follow the supervised learning paradigm\n", "* We have already seen for other structured prediction tasks\n", "* For supervised relation extraction, the scoring model \\\\(s_{\\params}(\\x,y)\\\\) is estimated automatically based on training sentences $\\mathcal{X}$ and their labels $\\mathcal{Y}$\n", "* We can use range of different classifiers, e.g. a logistic regression model or an SVM\n", "* At testing time, the predict label for each testing instance is the highest-scoring one, i.e. $$ \\y^* = \\argmax_{\\y\\in\\Ys} s(\\x,\\y) $$\n" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "* The training data consists again of patterns, entity pairs and labels\n", "* This time, the given labels for the training instances are 'method used for task' or 'NONE', i.e. we have positive and negative training data" ] }, { "cell_type": "code", "execution_count": 13, "metadata": { "slideshow": { "slide_type": "subslide" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Manually labelled data set consists of 22 negative training examples and 22 positive training examples\n", "\n", "demonstrates XXXXX and clustering techniques for XXXXX ['text mining', 'building domain ontology'] method used for task\n", "demonstrates text mining and XXXXX for building XXXXX ['clustering techniques', 'domain ontology'] method used for task\n", "the XXXXX is able to enhance the XXXXX ['ensemble classifier', 'detection of construction materials'] method used for task\n" ] } ], "source": [ "training_sents, training_entpairs, training_labels = ie.readLabelledData()\n", "print(\"Manually labelled data set consists of\", training_labels.count(\"NONE\"), \n", " \"negative training examples and\", training_labels.count(\"method used for task\"), \"positive training examples\\n\")\n", "for (tr_s, tr_e, tr_l) in zip(training_sents[:3], training_entpairs[:3], training_labels[:3]):\n", " print(tr_s, tr_e, tr_l)" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "* Next, we define how to transform training and testing data to features. \n", "* Features for the model are typically extracted from the shortest dependency path between two entities\n", "* Basic features are n-gram features, or they can be based on the syntactic structure of the input, i.e. the dependency path ([parsing](statnlpbook/chapters/parsing))\n", "* We assume again that entity pairs are part of the input, i.e. we assume the named entity recognition problem to be solved as part of the preprocessing of the data\n", "* In reality, named entities have to be recognised first\n", "\n", "* Let's look at an example, using one of sklearn's built-in feature extractor to transform sentences to n-grams" ] }, { "cell_type": "code", "execution_count": 14, "metadata": { "slideshow": { "slide_type": "subslide" } }, "outputs": [], "source": [ "from sklearn.feature_extraction.text import CountVectorizer\n", "\n", "def featTransform(sents_train, sents_test):\n", " cv = CountVectorizer()\n", " cv.fit(sents_train)\n", " print(cv.get_params())\n", " features_train = cv.transform(sents_train)\n", " features_test = cv.transform(sents_test)\n", " return features_train, features_test, cv" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "We define a model, again with sklearn, using one of their built-in classifiers and a prediction function." ] }, { "cell_type": "code", "execution_count": 15, "metadata": { "slideshow": { "slide_type": "subslide" } }, "outputs": [], "source": [ "from sklearn.linear_model import LogisticRegression\n", "\n", "def model_train(feats_train, labels):\n", " model = LogisticRegression(penalty='l2') # logistic regression model with l2 regularisation\n", " model.fit(feats_train, labels) # fit the model to the transformed training data\n", " return model\n", "\n", "def predict(model, features_test):\n", " \"\"\"Find the most compatible output class\"\"\"\n", " preds = model.predict(features_test) # this returns the predicted labels\n", " #preds_prob = model.predict_proba(features_test) # this returns probablities instead of labels\n", " return preds" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "We further define a helper function for debugging that determines the most useful features learned by the model" ] }, { "cell_type": "code", "execution_count": 16, "metadata": { "slideshow": { "slide_type": "subslide" } }, "outputs": [], "source": [ "def show_most_informative_features(vectorizer, clf, n=20):\n", " feature_names = vectorizer.get_feature_names()\n", " coefs_with_fns = sorted(zip(clf.coef_[0], feature_names))\n", " top = zip(coefs_with_fns[:n], coefs_with_fns[:-(n + 1):-1])\n", " for (coef_1, fn_1), (coef_2, fn_2) in top:\n", " print(\"\\t%.4f\\t%-15s\\t\\t%.4f\\t%-15s\" % (coef_1, fn_1, coef_2, fn_2))" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "Supervised relation extraction algorithm:" ] }, { "cell_type": "code", "execution_count": 17, "metadata": { "slideshow": { "slide_type": "subslide" } }, "outputs": [], "source": [ "def supervisedExtraction(train_sents, train_entpairs, train_labels, test_sents, test_entpairs):\n", " \"\"\"\n", " Given pos/neg training instances, train a logistic regression model with simple BOW features and predict labels on unseen test instances\n", " Args:\n", " train_sents: training sentences with arguments masked\n", " train_entpairs: training entity pairs\n", " train_labels: labels of training instances\n", " test_sents: testing sentences with arguments masked\n", " test_entpairs: testing entity pairs\n", " Returns:\n", " predictions for the testing sentences\n", " \"\"\"" ] }, { "cell_type": "code", "execution_count": 18, "metadata": { "slideshow": { "slide_type": "subslide" } }, "outputs": [], "source": [ "def supervisedExtraction(train_sents, train_entpairs, train_labels, test_sents, test_entpairs):\n", "\n", " # convert training and testing sentences to short paths to obtain patterns\n", " train_patterns = [sentenceToShortPath(test_sent) for test_sent in train_sents]\n", " test_patterns = [sentenceToShortPath(test_sent) for test_sent in test_sents]\n", "\n", " # extract features\n", " features_train, features_test, cv = featTransform(train_patterns, test_patterns)\n", "\n", " # train model\n", " model = model_train(features_train, train_labels)\n", "\n", " # show most common features\n", " show_most_informative_features(cv, model)\n", "\n", " # get predictions\n", " predictions = predict(model, features_test)\n", "\n", " # show the predictions\n", " for tup in zip(predictions, test_sents, test_entpairs):\n", " print(tup)\n", "\n", " return predictions" ] }, { "cell_type": "code", "execution_count": 19, "metadata": { "slideshow": { "slide_type": "subslide" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "{'analyzer': 'word', 'binary': False, 'decode_error': 'strict', 'dtype': , 'encoding': 'utf-8', 'input': 'content', 'lowercase': True, 'max_df': 1.0, 'max_features': None, 'min_df': 1, 'ngram_range': (1, 1), 'preprocessor': None, 'stop_words': None, 'strip_accents': None, 'token_pattern': '(?u)\\\\b\\\\w\\\\w+\\\\b', 'tokenizer': None, 'vocabulary': None}\n", "\t-0.9520\tof \t\t1.0542\tis \n", "\t-0.4352\tspecified \t\t0.9787\tto \n", "\t-0.4352\tusing \t\t0.8733\tfor \n", "\t-0.4313\tann \t\t0.4851\tand \n", "\t-0.4313\tfind \t\t0.4785\tsolved \n", "\t-0.3274\tdecreases \t\t0.4785\tassists \n", "\t-0.3258\tthat \t\t0.4309\tare \n", "\t-0.3181\tallowing \t\t0.4151\tsolve \n", "\t-0.3181\texcept \t\t0.4081\ton \n", "\t-0.3074\tas \t\t0.4081\tapplication \n", "\t-0.2935\tin \t\t0.3915\tmore \n", "\t-0.2892\tintroduced \t\t0.3915\tcapable \n", "\t-0.2892\tunified \t\t0.3526\tpresented \n", "\t-0.2755\tchecks \t\t0.3343\tadopted \n", "\t-0.2755\tcollision \t\t0.3208\tbuilding \n", "\t-0.2755\toverall \t\t0.3208\t3d \n", "\t-0.2755\tvirtual \t\t0.2736\tga \n", "\t-0.2715\tcan \t\t0.2736\tautomated \n", "\t-0.2715\tensure \t\t0.2398\tsizing \n", "\t-0.2651\talgorithms \t\t0.2240\tpso \n", "('NONE', 'a method for estimation of XXXXX of XXXXX is presented', ['effective properties', 'porous materials'])\n", "('method used for task', 'accounting for XXXXX is essential for estimation of XXXXX', ['nonlinear effects', 'effective properties'])\n", "('method used for task', 'develops the heterogeneous XXXXX for fiber-reinforced XXXXX', ['feature model', 'object modeling'])\n", "('NONE', 'two formulations for the problem of optimum XXXXX of onshore XXXXX', ['layout design', 'wind farms'])\n", "('method used for task', 'boundary-value and initial-value XXXXX are solved using XXXXX and graph products', ['differential equations', 'finite difference method'])\n", "('method used for task', 'boundary-value and initial-value XXXXX are solved using finite difference method and XXXXX', ['differential equations', 'graph products'])\n", "('method used for task', 'boundary-value and initial-value differential equations are solved using XXXXX and XXXXX', ['finite difference method', 'graph products'])\n", "('method used for task', 'an XXXXX to couple cfd and XXXXX is presented', ['open-source software', 'multiobjective optimization'])\n", "('method used for task', 'parallel XXXXX for solving cfd XXXXX is implemented', ['evolutionary algorithm', 'optimization problems'])\n", "('method used for task', 'a novel XXXXX platform is developed for the examination of an rfid-enabled mascs in a flexible XXXXX , and several system performance measures are considered in the XXXXX platform', ['assembly line', 'simulation test'])\n", "('method used for task', 'a novel simulation test platform is developed for the examination of an rfid-enabled mascs in a flexible XXXXX , and several XXXXX are considered in the simulation test platform', ['assembly line', 'system performance measures'])\n", "('method used for task', 'a novel XXXXX platform is developed for the examination of an rfid-enabled mascs in a flexible XXXXX , and several system performance measures are considered in the XXXXX platform', ['assembly line', 'simulation test'])\n", "('method used for task', 'a novel XXXXX platform is developed for the examination of an rfid-enabled mascs in a flexible assembly line , and several system performance measures are considered in the XXXXX platform', ['simulation test', 'simulation platform'])\n", "('method used for task', 'a novel XXXXX platform is developed for the examination of an rfid-enabled mascs in a flexible assembly line , and several XXXXX are considered in the XXXXX platform', ['simulation test', 'system performance measures'])\n", "('method used for task', 'a novel XXXXX platform is developed for the examination of an rfid-enabled mascs in a flexible assembly line , and several system performance measures are considered in the XXXXX platform', ['simulation test', 'simulation platform'])\n", "('method used for task', 'a novel XXXXX platform is developed for the examination of an rfid-enabled mascs in a flexible assembly line , and several system performance measures are considered in the XXXXX platform', ['simulation platform', 'simulation test'])\n", "('method used for task', 'a novel XXXXX platform is developed for the examination of an rfid-enabled mascs in a flexible assembly line , and several XXXXX are considered in the XXXXX platform', ['system performance measures', 'simulation test'])\n", "('method used for task', 'a novel XXXXX platform is developed for the examination of an rfid-enabled mascs in a flexible assembly line , and several system performance measures are considered in the XXXXX platform', ['simulation test', 'simulation platform'])\n", "('NONE', 'the XXXXX provides more efficient XXXXX', ['information exchange', 'reduced-order model server'])\n", "('method used for task', 'XXXXX and XXXXX that make the approach practical are presented', ['software design', 'implementation issues'])\n", "('NONE', 'the XXXXX of XXXXX can be generated , which can be used for the shape design of electrodes', ['cad model', 'volume feature'])\n", "('method used for task', 'the related parameters of XXXXX are extracted , which can be used for the XXXXX of electrodes and the setting of processing parameters', ['parametric design', 'volume feature'])\n", "('method used for task', 'an XXXXX is proposed for XXXXX of steel space frames under lrfd-aisc provisions', ['efficient algorithm', 'optimum design'])\n", "('method used for task', 'geometric nonlinear behavior of XXXXX is considered during the XXXXX', ['steel frames', 'optimization process'])\n", "('NONE', 'icde is extended to the XXXXX of XXXXX by combining icde with sora which gives a so-called the sora-icde', ['truss structures', 'rbdo problem'])\n", "('NONE', 'numerical results for five XXXXX illustrate the effectiveness of the sora-icde in solving the rbdo problem of XXXXX', ['benchmark problems', 'truss structures'])\n", "('NONE', 'numerical results for five XXXXX illustrate the effectiveness of the sora-icde in solving the XXXXX of truss structures', ['benchmark problems', 'rbdo problem'])\n", "('NONE', 'numerical results for five benchmark problems illustrate the effectiveness of the sora-icde in solving the XXXXX of XXXXX', ['truss structures', 'rbdo problem'])\n", "('NONE', 'XXXXX of the wheel is estimated using stress life ( s-n ) method based on the XXXXX of the wheel', ['fatigue life', 'stress analysis'])\n", "('NONE', 'simulations of ballistic XXXXX were performed using the validated XXXXX', ['impact tests', 'fe model'])\n", "('NONE', 'static and XXXXX of variable geometry microbeams using XXXXX', ['dynamic analysis', 'energy method'])\n", "('method used for task', 'a XXXXX is proposed to handle thin-walled model with XXXXX', ['hybrid method', 'complex geometry'])\n", "('method used for task', 'the XXXXX of woa is confirmed by the results on XXXXX', ['multimodal functions', 'exploration ability'])\n", "('method used for task', 'this paper develops large scale marine hydrological environmental data-oriented XXXXX and realizes oceanographic planar graph , contour line rendering , isosurface rendering , factor field volume rendering and XXXXX of current field', ['dynamic simulation', 'visualization software'])\n", "('NONE', 'the system employs cuda XXXXX to improve the computation rate of XXXXX of marine water environmental factors based on netcdf ( network common data form ) format', ['parallel computing', 'volume rendering'])\n", "('method used for task', 'the system employs cuda XXXXX to improve the XXXXX of volume rendering of marine water environmental factors based on netcdf ( network common data form ) format', ['parallel computing', 'computation rate'])\n", "('NONE', 'the system employs cuda parallel computing to improve the XXXXX of XXXXX of marine water environmental factors based on netcdf ( network common data form ) format', ['volume rendering', 'computation rate'])\n", "('method used for task', 'nuclear XXXXX modeling generality and robustness are improved by a modular , agent based XXXXX', ['modeling framework', 'fuel cycle'])\n", "('NONE', 'paper reviews applications of XXXXX in XXXXX', ['artificial neural networks', 'model calibration'])\n", "('method used for task', 'worst XXXXX oriented approach for volume fraction minimization in XXXXX with uncertain load directions and compliance constraints is considered', ['topology optimization', 'load direction'])\n", "('method used for task', 'the stored data from XXXXX are replaced by continuous functions suitable for representation in XXXXX', ['finite element analysis', 'computer graphics'])\n", "('method used for task', 'the approach is based on a three-dimensional XXXXX and requires a low XXXXX', ['discrete element', 'computational cost'])\n", "('method used for task', 'the system is based on XXXXX and XXXXX', ['artificial neural networks', 'genetic algorithm'])\n", "('NONE', 'XXXXX for estimation of XXXXX', ['soft computing methods', 'wake wind speed'])\n", "('method used for task', 'propose the method of setting basic constraints to solve the shortcomings when XXXXX and haptic feedback are both integrated in assembly tasks . the basic constraints make the XXXXX easy and realistic through the usage of haptics and visual fidelity', ['physics engine', 'assembly training'])\n", "('NONE', 'except the display and XXXXX modules , the whole XXXXX is made up of closeable widgets', ['physics engine', 'training system'])\n", "('NONE', 'XXXXX of the XXXXX to mpi-only parallelization', ['performance comparison', 'hybrid parallelization'])\n", "('NONE', 'it is advisable among dentists to perform as diverse XXXXX as possible to reduce the risk of decreased XXXXX', ['work tasks', 'pinch grip strength'])\n", "('method used for task', 'delivering a first XXXXX for designing sustainable XXXXX', ['frame work', 'work systems'])\n", "('method used for task', 'a XXXXX function was derived to reveal the relationship between the muscle load and the XXXXX', ['subjective evaluation', 'satisfaction level'])\n", "('method used for task', 'a satisfaction level function was derived to reveal the relationship between the XXXXX and the XXXXX', ['subjective evaluation', 'muscle load'])\n", "('method used for task', 'a XXXXX function was derived to reveal the relationship between the XXXXX and the subjective evaluation', ['satisfaction level', 'muscle load'])\n", "('NONE', 'evaluation of on-road XXXXX was performed using XXXXX on 12 male volunteers', ['surface emg', 'bicycle design'])\n", "('method used for task', 'we formulate the relationship between perceived discomfort and joint XXXXX for the XXXXX', ['upper limb', 'moment ratio'])\n", "('method used for task', 'reconfiguration reduced XXXXX and enhanced XXXXX', ['error rates', 'user acceptance'])\n", "('method used for task', 'the seips model of XXXXX and XXXXX is a useful systems approach to healthcare quality and XXXXX', ['work system', 'patient safety'])\n", "('method used for task', 'the seips model of XXXXX and patient safety is a useful XXXXX to healthcare quality and patient safety', ['work system', 'systems approach'])\n", "('method used for task', 'the seips model of XXXXX and patient safety is a useful systems approach to XXXXX and patient safety', ['work system', 'healthcare quality'])\n", "('method used for task', 'the seips model of XXXXX and XXXXX is a useful systems approach to healthcare quality and XXXXX', ['work system', 'patient safety'])\n", "('method used for task', 'the seips model of work system and XXXXX is a useful XXXXX to healthcare quality and XXXXX', ['patient safety', 'systems approach'])\n", "('method used for task', 'the seips model of work system and XXXXX is a useful systems approach to XXXXX and XXXXX', ['patient safety', 'healthcare quality'])\n", "('method used for task', 'the seips model of work system and patient safety is a useful XXXXX to XXXXX and patient safety', ['systems approach', 'healthcare quality'])\n", "('method used for task', 'the seips model of work system and XXXXX is a useful XXXXX to healthcare quality and XXXXX', ['systems approach', 'patient safety'])\n", "('method used for task', 'the seips model of work system and XXXXX is a useful systems approach to XXXXX and XXXXX', ['healthcare quality', 'patient safety'])\n", "('method used for task', 'the seips model can be used for research and improvement activities for improving XXXXX and XXXXX', ['healthcare quality', 'patient safety'])\n", "('method used for task', 'the seips model can be used for research and XXXXX for improving XXXXX and patient safety', ['healthcare quality', 'improvement activities'])\n", "('method used for task', 'the seips model can be used for research and XXXXX for improving healthcare quality and XXXXX', ['patient safety', 'improvement activities'])\n", "('method used for task', 'balancing the XXXXX is a key principle to improve XXXXX and patient safety', ['work system', 'healthcare quality'])\n", "('method used for task', 'balancing the XXXXX is a key principle to improve healthcare quality and XXXXX', ['work system', 'patient safety'])\n", "('method used for task', 'balancing the work system is a key principle to improve XXXXX and XXXXX', ['healthcare quality', 'patient safety'])\n", "('NONE', 'an intergroup XXXXX identified distinct XXXXX of home heating that explained users reported behaviour', ['case study', 'mental models'])\n", "('NONE', 'an intergroup XXXXX identified distinct mental models of XXXXX that explained users reported behaviour', ['case study', 'home heating'])\n", "('NONE', 'an intergroup case study identified distinct XXXXX of XXXXX that explained users reported behaviour', ['mental models', 'home heating'])\n", "('NONE', 'electromyography , XXXXX , and ratings of XXXXX differentiated 3 hand-carried devices and a manual carry', ['heart rate', 'perceived exertion'])\n", "('method used for task', 'guidelines for developing a method for assessing all XXXXX and all XXXXX', ['work tasks', 'body parts'])\n", "('method used for task', 'wearing the XXXXX and s.c.b.a. , 14 % of all firefighters recruits failed to complete the XXXXX', ['protective clothing', 'endurance test'])\n", "('NONE', 'in a XXXXX ergonomic XXXXX are likely to be transformed , not just transferred', ['design process', 'guideline texts'])\n", "('NONE', \"XXXXX task 's ( lct ) sensitivity towards XXXXX was examined\", ['learning effects', 'lane change'])\n", "('method used for task', 'camouflage effectiveness was evaluated using hit rate , XXXXX , and XXXXX', ['detection time', 'eye movement data'])\n", "('NONE', 'results suggest that sa XXXXX showed some XXXXX', ['common ground', 'information elements'])\n", "('method used for task', 'trapezius XXXXX and workload were assessed over full day and XXXXX in nurses', ['muscle activity', 'night shifts'])\n", "('NONE', 'during day and XXXXX , XXXXX was very high compared to resting values', ['heart rate', 'night shifts'])\n", "('method used for task', 'perception of XXXXX and mental well-being at work were similar during day and XXXXX', ['neck pain', 'night shifts'])\n", "('NONE', 'XXXXX were perceived to be as burdensome as day shifts despite the smaller XXXXX', ['physical workload', 'night shifts'])\n", "('NONE', 'tested 14 universal XXXXX in the XXXXX , korea , and turkey', ['united states', 'healthcare symbols'])\n", "('NONE', 'poor health and XXXXX did not interact regarding reduced demand-specific XXXXX', ['working conditions', 'work ability'])\n", "('method used for task', 'XXXXX location varies by vehicle and XXXXX', ['hand placement', 'age group'])\n", "('NONE', 'effect of XXXXX predictability on XXXXX task ( lct ) performance was tested', ['lane change', 'lane task'])\n", "('NONE', 'effect of XXXXX predictability on XXXXX task ( lct ) performance was tested', ['lane change', 'lane task'])\n", "('NONE', 'finger flexor and extensor XXXXX was lower during the virtual XXXXX', ['muscle activity', 'keyboard use'])\n", "('NONE', 'a novel approach was developed to determine the XXXXX in XXXXX', ['air gap', 'protective clothing'])\n", "('method used for task', 'the XXXXX depended on the XXXXX , fabric properties , and garment size', ['air gap', 'body parts'])\n", "('NONE', 'neither system reduced core temperature , XXXXX , XXXXX or comfort', ['heart rate', 'perceived exertion'])\n", "('NONE', 'neither system reduced XXXXX , XXXXX , perceived exertion or comfort', ['heart rate', 'core temperature'])\n", "('NONE', 'neither system reduced XXXXX , heart rate , XXXXX or comfort', ['perceived exertion', 'core temperature'])\n", "('NONE', 'XXXXX in and around the firehouse may increase XXXXX', ['physical activity', 'body temperature'])\n", "('NONE', 'a list of XXXXX that influenced trust was derived from XXXXX', ['qualitative analysis', 'system features'])\n", "('NONE', 'parents and XXXXX members who reviewed video records of their bedside rounds participated in the analysis of XXXXX and facilitators in family-centered rounds', ['healthcare team', 'work system barriers'])\n", "('method used for task', 'the stimulated XXXXX was positively received by parents and XXXXX members', ['recall methodology', 'healthcare team'])\n", "('NONE', 'the stimulated XXXXX allowed the identification of a wide range of XXXXX and facilitators in family-centered rounds', ['recall methodology', 'work system barriers'])\n", "('NONE', 'XXXXX significantly lower on XXXXX due to decreased performance', ['muscle activity', 'slate computer'])\n", "('NONE', 'a toolkit approach is proposed to improve XXXXX of wmsds in XXXXX', ['risk management', 'health care'])\n", "('NONE', 'four different types of insoles were examined in terms of their effects on XXXXX in XXXXX', ['postural stability', 'older adults'])\n", "('NONE', 'the findings can aid in better understanding the insole XXXXX that could improve XXXXX in older adults', ['design features', 'postural stability'])\n", "('NONE', 'the findings can aid in better understanding the insole XXXXX that could improve postural stability in XXXXX', ['design features', 'older adults'])\n", "('NONE', 'the findings can aid in better understanding the insole design features that could improve XXXXX in XXXXX', ['postural stability', 'older adults'])\n", "('NONE', 'the effect of electric XXXXX on human heat balance during XXXXX was modelled', ['fan use', 'heat waves'])\n", "('NONE', 'we performed XXXXX of assembly training for automotive XXXXX', ['user studies', 'assembly lines'])\n", "('NONE', 'we performed XXXXX of XXXXX for automotive assembly lines', ['user studies', 'assembly training'])\n", "('method used for task', 'we performed user studies of XXXXX for automotive XXXXX', ['assembly lines', 'assembly training'])\n", "('method used for task', 'the study describes a longitudinal qualitative XXXXX examining efforts to improve behavioural XXXXX over a two- year period', ['case study', 'energy efficiency'])\n", "('method used for task', 'an association between prolonged XXXXX and XXXXX was found among women', ['arm elevation', 'shoulder pain'])\n", "('NONE', 'XXXXX increases carpal XXXXX in patients with carpal tunnel syndrome', ['computer use', 'tunnel pressure'])\n", "('method used for task', 'XXXXX times for XXXXX are slightly higher than for guide signs', ['visual processing', 'logo signs'])\n", "('method used for task', 'increased XXXXX times for XXXXX do not result in any vehicle control degradation', ['visual processing', 'logo signs'])\n", "('NONE', \"patients ' and informal caregivers ' XXXXX was shaped by XXXXX\", ['work performance', 'system factors'])\n", "('NONE', 'the XXXXX of the XXXXX has been calculated', ['human error probability', 'case system'])\n", "('NONE', 'some strategies to prevent latent XXXXX of the XXXXX have been proposed', ['human errors', 'case system'])\n", "('NONE', 'presentation of a personal XXXXX ( pcs ) XXXXX', ['cooling system', 'selection method'])\n", "('method used for task', 'the new XXXXX was superior to the current XXXXX in XXXXX and satisfaction score', ['error rate', 'touch method'])\n", "('method used for task', 'the new XXXXX was superior to the current XXXXX in XXXXX and satisfaction score', ['error rate', 'touch method'])\n", "('method used for task', 'XXXXX for the XXXXX have the least effect on the prediction', ['measurement errors', 'arc length'])\n", "('NONE', \"XXXXX of sports t-shirt increases wearer 's XXXXX\", ['thermal comfort', 'ventilation design'])\n", "('method used for task', 'an XXXXX on the human XXXXX inside the aircrafts was carried out', ['experimental investigation', 'thermal comfort'])\n", "('method used for task', 'appropriate ud XXXXX are varied on XXXXX', ['evaluation scales', 'product attributes'])\n", "('NONE', 'written XXXXX resulted in more accurate XXXXX than spoken channels', ['communication channels', 'message transmission'])\n", "('method used for task', 'high XXXXX forces and plantar pressures are observed during XXXXX at high gait cadences', ['load carriage', 'ground reaction'])\n", "('method used for task', 'the metabolic XXXXX was found to have a basic impact on the XXXXX', ['heat generation', 'temperature profile'])\n", "('method used for task', 'XXXXX was quantified and examined in relation to XXXXX', ['physical workload', 'strength capacities'])\n", "('NONE', 'XXXXX , XXXXX , trunk motion , and perceived exertion data were collected', ['surface emg', 'heart rate'])\n", "('method used for task', 'we present a practical approach based on XXXXX to estimate XXXXX using heart rate monitoring', ['neuro-fuzzy systems', 'energy expenditure'])\n", "('NONE', 'XXXXX texting requires less XXXXX than a physical keypad', ['touch screen', 'muscle activity'])\n", "('NONE', 'XXXXX demonstrates a more dynamic spine than XXXXX', ['keyboard use', 'mouse use'])\n", "('method used for task', 'speech-based agents enhanced driver XXXXX and XXXXX', ['situation awareness', 'driving performance'])\n", "('NONE', \"a driver 's XXXXX mediated anger effects on XXXXX\", ['situation awareness', 'driving performance'])\n", "('method used for task', 'we assess footwear XXXXX on ice by the maximum XXXXX subjects could stand and walk', ['slip resistance', 'slope angles'])\n", "('NONE', 'the maximum XXXXX were objective and ecologically valid measures of XXXXX', ['slip resistance', 'slope angles'])\n", "('NONE', \"this human-centred measure of XXXXX did not require controlling of individuals ' XXXXX\", ['slip resistance', 'gait characteristics'])\n", "('NONE', 'we conducted a XXXXX experiment under real XXXXX', ['lane change', 'road environment'])\n", "('NONE', 'XXXXX intent XXXXX is about 5 s', ['lane changing', 'time window'])\n", "('method used for task', 'vehicle motion states , driving conditions and XXXXX information were chosen to predict XXXXX behaviours', ['lane changing', 'head movements'])\n", "('NONE', 'the improved XXXXX detects 85 % of XXXXX 1.5 s in advance', ['neural network', 'lane changes'])\n", "('NONE', 'XXXXX that enriches how physicians view XXXXX', ['web-based training', 'diagnosis delivery'])\n", "('NONE', 'XXXXX provides XXXXX during blindfolded water pouring', ['motion capture', 'auditory feedback'])\n", "('method used for task', 'XXXXX and XXXXX are provided based on these results', ['system improvement', 'design recommendations'])\n", "('method used for task', 'we propose a novel hybrid XXXXX able to handle XXXXX sizes', ['small sample', 'regression method'])\n", "('NONE', 'outlines role of XXXXX in sustainability , and the role of XXXXX in these systems', ['cyber-physical systems', 'systems ergonomics'])\n", "('NONE', 'we examine XXXXX as a mediator in the relationships between job characteristics and XXXXX', ['job engagement', 'safety performance'])\n", "('method used for task', 'job resources had positive effects on XXXXX directly and indirectly through XXXXX', ['safety performance', 'job engagement'])\n", "('NONE', 'XXXXX can translate into better XXXXX', ['employee engagement', 'safety performance'])\n", "('NONE', 'XXXXX of XXXXX defined', ['objective measures', 'tool use'])\n", "('method used for task', 'XXXXX for site-specific schoolbag-related XXXXX were identified', ['risk factors', 'musculoskeletal discomfort'])\n", "('method used for task', 'we examine the influence of XXXXX and driving experience on subjective workload and XXXXX', ['driving performance', 'situation complexity'])\n", "('method used for task', 'we used XXXXX to analyse rail XXXXX', ['cognitive work analysis', 'level crossings'])\n", "('NONE', 'effects of individual and XXXXX on XXXXX were investigated', ['grip strength', 'job factors'])\n", "('NONE', 'opportunities exist to optimise XXXXX using XXXXX', ['virtual reality', 'vehicle development'])\n", "('NONE', 'the XXXXX reduced seat back XXXXX at the low back', ['thoracic support', 'contact pressure area'])\n", "('NONE', 'XXXXX played a leading causal role in clinical incidents , with XXXXX varying by type of action', ['human error', 'error type'])\n", "('NONE', 'it is suggested to evaluate the overall XXXXX of passengers using one single XXXXX ranging from extreme discomfort to extreme comfort', ['comfort experience', 'rating scale'])\n", "('method used for task', 'the XXXXX and XXXXX literatures are reviewed for insights regarding effective decision support design', ['decision making', 'problem solving'])\n", "('method used for task', 'the XXXXX and problem solving literatures are reviewed for insights regarding effective XXXXX', ['decision making', 'decision support design'])\n", "('method used for task', 'the decision making and XXXXX literatures are reviewed for insights regarding effective XXXXX', ['problem solving', 'decision support design'])\n", "('NONE', 'the role of graphical displays in supporting XXXXX , XXXXX and system safety is explored in detail', ['decision making', 'problem solving'])\n", "('method used for task', 'the role of graphical displays in supporting XXXXX , problem solving and XXXXX is explored in detail', ['decision making', 'system safety'])\n", "('method used for task', 'the role of graphical displays in supporting decision making , XXXXX and XXXXX is explored in detail', ['problem solving', 'system safety'])\n", "('NONE', 'participants prefer auditory warnings , as opposed to visual , vibrotactile or electric stimuli . specifically , XXXXX prefer alarm warnings , while XXXXX prefer verbal warnings', ['truck drivers', 'taxi drivers'])\n", "('NONE', 'XXXXX via the combination of technologies incorporating XXXXX , tri-axial accelerometers , and supported by survey data', ['data collection', 'gps data'])\n", "('NONE', 'XXXXX via the combination of technologies incorporating gps data , tri-axial accelerometers , and supported by XXXXX', ['data collection', 'survey data'])\n", "('method used for task', 'data collection via the combination of technologies incorporating XXXXX , tri-axial accelerometers , and supported by XXXXX', ['gps data', 'survey data'])\n", "('NONE', 'the maximum XXXXX and fatigue were influenced significantly in XXXXX and pressure', ['grip strength', 'low temperature'])\n", "('NONE', 'work sampling , XXXXX , XXXXX were integrated', ['computer simulation', 'biomechanical modeling'])\n", "('NONE', 'intrinsic XXXXX is an essential feature of XXXXX', ['human motion', 'movement variability'])\n", "('NONE', 'XXXXX theories can explain the existence of intrinsic XXXXX', ['motor control', 'movement variability'])\n", "('NONE', 'intrinsic XXXXX should be taken into account in XXXXX', ['movement variability', 'workstation design'])\n", "('method used for task', 'XXXXX and XXXXX in grocery store workers are clarified', ['physical workload', 'musculoskeletal symptoms'])\n", "('method used for task', 'XXXXX ( cwa ) is applied to pedestrian use of XXXXX', ['cognitive work analysis', 'rail level crossings'])\n", "('NONE', 'XXXXX was influenced by XXXXX but not by lay-off period', ['system reliability', 'operator trust'])\n", "('method used for task', 'XXXXX climate perceptions were linked to XXXXX and engagement', ['job satisfaction', 'employee safety'])\n", "('method used for task', 'XXXXX climate perceptions were linked to objective XXXXX', ['employee safety', 'turnover rate'])\n", "('NONE', 'XXXXX mediated between XXXXX , employee engagement , turnover rate', ['job satisfaction', 'safety climate'])\n", "('NONE', 'XXXXX mediated between safety climate , XXXXX , turnover rate', ['job satisfaction', 'employee engagement'])\n", "('NONE', 'XXXXX mediated between safety climate , employee engagement , XXXXX', ['job satisfaction', 'turnover rate'])\n", "('NONE', 'job satisfaction mediated between XXXXX , XXXXX , turnover rate', ['safety climate', 'employee engagement'])\n", "('NONE', 'job satisfaction mediated between XXXXX , employee engagement , XXXXX', ['safety climate', 'turnover rate'])\n", "('NONE', 'job satisfaction mediated between safety climate , XXXXX , XXXXX', ['employee engagement', 'turnover rate'])\n", "('method used for task', 'XXXXX should focus on the interactions between XXXXX and the role of higher level work system factors', ['future research', 'work systems'])\n", "('NONE', 'we examine whether XXXXX ( wbv ) increases frequency of XXXXX', ['whole-body vibration', 'action slips'])\n", "('method used for task', 'XXXXX times didn’t differ between 2d and 3d , but times were slower on the XXXXX than the microscope', ['task completion', 'video displays'])\n", "('method used for task', 'XXXXX reduce posture constraints and may reduce XXXXX and fatigue in microsurgery', ['musculoskeletal symptoms', 'video displays'])\n", "('NONE', 'the need of approaching and intervening simultaneously human and environmental problems has been identify . this are the results of a XXXXX of design concepts and methods associated with human and XXXXX', ['systematic review', 'environmental factors'])\n", "('NONE', 'the need of approaching and intervening simultaneously human and environmental problems has been identify . this are the results of a XXXXX of XXXXX and methods associated with human and environmental factors', ['systematic review', 'design concepts'])\n", "('method used for task', 'the need of approaching and intervening simultaneously human and environmental problems has been identify . this are the results of a systematic review of XXXXX and methods associated with human and XXXXX', ['environmental factors', 'design concepts'])\n", "('method used for task', 'it shows conceptual and methodological segregation between XXXXX and XXXXX in published documents', ['human factors', 'environmental factors'])\n", "('NONE', 'XXXXX was utilized to assess the improvement strategy of latent XXXXX', ['fuzzy topsis', 'human errors'])\n", "('NONE', 'XXXXX was used to inspect the robustness of the results of XXXXX', ['sensitivity analysis', 'fuzzy topsis'])\n", "('method used for task', 'this is the first study to assess XXXXX by hfacs and XXXXX', ['human errors', 'fuzzy topsis'])\n", "('method used for task', 'XXXXX are neither necessary nor sufficient to elicit XXXXX', ['time constraints', 'time pressure'])\n", "('method used for task', 'sorting quality influences XXXXX , environment and XXXXX', ['working conditions', 'system performance'])\n", "('method used for task', 'maritime engine room officers commonly agree that the rapid XXXXX and reduced staffing onboard have contributed to higher workload and altered XXXXX', ['technological development', 'work tasks'])\n", "('NONE', 'presented and provided evidence of how the XXXXX are categorised in each XXXXX', ['visualisation methods', 'selection approach'])\n", "('method used for task', 'neck disorders were associated with head and XXXXX , trapezius and forearm XXXXX , and wrist velocity', ['muscle activity', 'arm posture'])\n", "('NONE', 'workstations on XXXXX were classified into high and low workload workstations with the median score of XXXXX', ['assembly line', 'body parts'])\n", "('NONE', 'the hlcb could lower XXXXX , oxygen uptake , minute ventilation and peripheral rating of XXXXX when carrying a 15 kg load', ['heart rate', 'perceived exertion'])\n", "('NONE', 'XXXXX can address XXXXX ( e.g. , enhancing awareness )', ['human factors', 'policy interventions'])\n", "('NONE', 'the decision ladder was used to examine XXXXX at rail XXXXX', ['decision making', 'level crossings'])\n", "('NONE', 'XXXXX at rail XXXXX varies within and between road user groups', ['decision making', 'level crossings'])\n", "('NONE', 'XXXXX sonifications do not reveal when XXXXX is too high', ['pulse oximetry', 'oxygen saturation'])\n", "('NONE', 'pain reports of those with XXXXX on XXXXX decreased 58 % on decline', ['low back pain', 'level ground'])\n", "('NONE', 'during an instrument XXXXX proficiency test , XXXXX ( hr ) /heart rate variation ( hrv ) are sensitive measures of varying levels of pilot mental workload', ['heart rate', 'flight rules'])\n", "('NONE', 'this new approach provides more degrees of freedom and XXXXX in XXXXX', ['type-2 fuzzy logic systems', 'design flexibility'])\n", "('NONE', 'XXXXX using granular type-2 XXXXX have more potential to model and handle uncertainties', ['type-2 fuzzy logic systems', 'membership functions'])\n", "('NONE', 'XXXXX has been carried out without removing XXXXX in the data set', ['feature selection', 'missing values'])\n", "('method used for task', 'we show the application of gifss in real XXXXX and XXXXX', ['supplier selection', 'medical diagnosis problems'])\n", "('NONE', 'a matheuristic by combining the XXXXX with a XXXXX is suggested', ['tabu search algorithm', 'neighbourhood structure'])\n", "('NONE', 'we make use of XXXXX as a XXXXX for illumination classification', ['rough set', 'decision maker'])\n", "('method used for task', 'the evolving XXXXX is introduced for the modelling of XXXXX with dead-zone input', ['intelligent algorithm', 'nonlinear systems'])\n", "('NONE', 'applying the evolving computation concept to train a dynamic XXXXX capable of modeling XXXXX alloy actuators', ['black box', 'shape memory'])\n", "('NONE', 'we employ partially XXXXX for solving system of XXXXX', ['fuzzy neural network', 'fuzzy differential equations'])\n", "('NONE', 'we propose a XXXXX from the XXXXX for adjusting of weights', ['learning algorithm', 'cost function'])\n", "('NONE', 'XXXXX of open XXXXX and state of charge of lifepo4 batteries', ['mathematical model', 'circuit voltage'])\n", "('method used for task', 'use of XXXXX and fourier XXXXX for identifying similar past situations', ['hierarchical clustering', 'frequency analysis'])\n", "('method used for task', 'we use XXXXX to handle the XXXXX in the real-world problems', ['fuzzy set theory', 'imprecise information'])\n", "('method used for task', 'we use XXXXX to handle the imprecise information in the XXXXX', ['fuzzy set theory', 'real-world problems'])\n", "('NONE', 'we use fuzzy set theory to handle the XXXXX in the XXXXX', ['imprecise information', 'real-world problems'])\n", "('method used for task', 'a binary-real-coded XXXXX ( ga ) is proposed as the XXXXX of the ucp', ['genetic algorithm', 'solution technique'])\n", "('method used for task', 'we have considered in the XXXXX : continuous , integer and binary XXXXX simultaneously', ['optimization problem', 'decision variables'])\n", "('NONE', 'we introduce and examine a new methodology for training XXXXX ( rbf ) XXXXX', ['radial basis function', 'neural networks'])\n", "('NONE', 'we use XXXXX in order to improve the functionality of the optimum XXXXX ( osd ) learning algorithm', ['fuzzy clustering', 'steepest descent'])\n", "('NONE', 'we use XXXXX in order to improve the functionality of the optimum steepest descent ( osd ) XXXXX', ['fuzzy clustering', 'learning algorithm'])\n", "('NONE', 'we use fuzzy clustering in order to improve the functionality of the optimum XXXXX ( osd ) XXXXX', ['steepest descent', 'learning algorithm'])\n", "('method used for task', 'we can synthesize a set of XXXXX and find appropriate dithers to stabilize nonlinear multiple time-delay ( nmtd ) XXXXX', ['fuzzy controllers', 'interconnected systems'])\n", "('NONE', 'when the designed XXXXX can not stabilize the nmtd XXXXX , a batch of high-frequency signals ( commonly referred to as dithers ) is simultaneously introduced to stabilize it', ['fuzzy controllers', 'interconnected systems'])\n", "('NONE', 'gea-based pso can perform a XXXXX with faster XXXXX', ['global search', 'convergence speed'])\n", "('NONE', 'a hybrid optimization method for XXXXX of XXXXX in distribution networks is proposed', ['optimal allocation', 'wind turbines'])\n", "('method used for task', 'a XXXXX for XXXXX of wind turbines in distribution networks is proposed', ['optimal allocation', 'hybrid optimization method'])\n", "('NONE', 'a XXXXX for optimal allocation of XXXXX in distribution networks is proposed', ['wind turbines', 'hybrid optimization method'])\n", "('method used for task', 'combining the XXXXX ( ga ) and the market-based XXXXX ( opf )', ['genetic algorithm', 'optimal power flow'])\n", "('method used for task', 'diesel engine models are built using advanced XXXXX and verified based on XXXXX', ['machine learning techniques', 'experimental data'])\n", "('NONE', 'ntld classifier gives good XXXXX in case of XXXXX', ['classification accuracy', 'multi-class problem'])\n", "('method used for task', 'the solutions delivered by vn-mga are compared with the ones delivered by the XXXXX provided by an ilp solver , for medium-sized problem instances . the vn-mga was found to be able to find most of the exact XXXXX , leaving a small gap in the other cases', ['exact solutions', 'pareto-optimal solutions'])\n", "('NONE', 'the solutions delivered by vn-mga are compared with the ones delivered by the XXXXX provided by an XXXXX , for medium-sized problem instances . the vn-mga was found to be able to find most of the exact pareto-optimal solutions , leaving a small gap in the other cases', ['exact solutions', 'ilp solver'])\n", "('method used for task', 'the solutions delivered by vn-mga are compared with the ones delivered by the XXXXX provided by an ilp solver , for medium-sized XXXXX . the vn-mga was found to be able to find most of the exact pareto-optimal solutions , leaving a small gap in the other cases', ['exact solutions', 'problem instances'])\n", "('method used for task', 'the solutions delivered by vn-mga are compared with the ones delivered by the exact solutions provided by an XXXXX , for medium-sized problem instances . the vn-mga was found to be able to find most of the exact XXXXX , leaving a small gap in the other cases', ['pareto-optimal solutions', 'ilp solver'])\n", "('method used for task', 'the solutions delivered by vn-mga are compared with the ones delivered by the exact solutions provided by an ilp solver , for medium-sized XXXXX . the vn-mga was found to be able to find most of the exact XXXXX , leaving a small gap in the other cases', ['pareto-optimal solutions', 'problem instances'])\n", "('method used for task', 'the solutions delivered by vn-mga are compared with the ones delivered by the exact solutions provided by an XXXXX , for medium-sized XXXXX . the vn-mga was found to be able to find most of the exact pareto-optimal solutions , leaving a small gap in the other cases', ['ilp solver', 'problem instances'])\n", "('NONE', 'we present a supplier selection decision method based on XXXXX combined with a fuzzy rule-based XXXXX', ['fuzzy inference', 'classification method'])\n", "('method used for task', 'an adaptive hysteresis current XXXXX is proposed for dc–ac inverter in the XXXXX', ['control algorithm', 'pv system'])\n", "('NONE', 'we propose a network clustering algorithm , based on the application of XXXXX and capable of exploiting the XXXXX', ['genetic operators', 'traffic information'])\n", "('NONE', 'biological motivation , XXXXX , XXXXX , open research problems and challenging issues of these models', ['design principles', 'application areas'])\n", "('NONE', 'distributed XXXXX in a XXXXX including v2g', ['energy resources', 'smart grid environment'])\n", "('method used for task', 'an XXXXX named mofoa is proposed , which is the first study that extends the relatively new fireworks optimization heuristic for XXXXX', ['evolutionary algorithm', 'multiobjective optimization'])\n", "('NONE', 'XXXXX : modified shuffled frog leaping algorithm ( msfla ) with XXXXX ( ga ) cross-over', ['genetic algorithm', 'solution method'])\n", "('method used for task', 'an improved variant of the XXXXX is presented by simulating a XXXXX', ['abc algorithm', 'learning mechanism'])\n", "('NONE', 'indicating the XXXXX problems of the analytic XXXXX ( ahp ) , and proposing the paired interval scale addressing the limitations', ['rating scale', 'hierarchy process'])\n", "('NONE', 'a XXXXX combining XXXXX with wavelet radial basis function neural network is presented', ['hybrid algorithm', 'particle swarm optimization'])\n", "('NONE', 'a XXXXX combining particle swarm optimization with wavelet XXXXX is presented', ['hybrid algorithm', 'radial basis function neural network'])\n", "('NONE', 'a hybrid algorithm combining XXXXX with wavelet XXXXX is presented', ['particle swarm optimization', 'radial basis function neural network'])\n", "('NONE', 'development of evolutionary-tuned XXXXX for dynamic sequencing of jobs on a XXXXX', ['fuzzy controller', 'manufacturing facility'])\n", "('NONE', 'the proposed controllers produced considerably more XXXXX than the examined XXXXX', ['pareto optimal solutions', 'dispatching rules'])\n", "('method used for task', 'we propose XXXXX which is based on combining adaboost algorithm with svm for XXXXX phenomenon . we call this boosted svm', ['hybrid approach', 'imbalanced data'])\n", "('NONE', 'the proposed XXXXX minimizes weighted exponential XXXXX', ['hybrid approach', 'error function'])\n", "('NONE', 'the problem of constructing an organ XXXXX via XXXXX is assessed', ['machine learning techniques', 'allocation model'])\n", "('method used for task', 'the conclusion is based on the XXXXX and XXXXX ( i.e. , wilcoxon signed rank median test )', ['performance metrics', 'statistical analysis'])\n", "('method used for task', 'XXXXX for finite state machines is one of the main XXXXX in the synthesis of sequential circuits', ['state assignment', 'optimization problems'])\n", "('NONE', 'XXXXX for finite state machines is one of the main optimization problems in the synthesis of XXXXX', ['state assignment', 'sequential circuits'])\n", "('NONE', 'state assignment for finite state machines is one of the main XXXXX in the synthesis of XXXXX', ['optimization problems', 'sequential circuits'])\n", "('method used for task', 'an improved binary XXXXX is proposed and its effectiveness is demonstrated in solving the XXXXX', ['pso algorithm', 'state assignment problem'])\n", "('method used for task', 'one of the objective of XXXXX is to synthesize XXXXX targeting area optimization', ['sequential circuits', 'state assignment problem'])\n", "('NONE', 'an XXXXX using itae , XXXXX and settling times is proposed', ['objective function', 'damping ratio'])\n", "('NONE', 'an XXXXX for training the parameters of the XXXXX is further developed', ['optimal algorithm', 'forecasting model'])\n", "('method used for task', 'a XXXXX is examined to demonstrate the ability of the newly proposed XXXXX', ['case study', 'forecasting model'])\n", "('method used for task', 'the XXXXX approach is used to optimize local and global modelling capability of XXXXX', ['t-s fuzzy model', 'weighting parameters'])\n", "('NONE', 'the weighting parameters approach is used to optimize local and global XXXXX of XXXXX', ['t-s fuzzy model', 'modelling capability'])\n", "('method used for task', 'the XXXXX approach is used to optimize local and global XXXXX of t-s fuzzy model', ['weighting parameters', 'modelling capability'])\n", "('method used for task', 'proposed a XXXXX to classify the XXXXX images into two outcomes : benign or malignant', ['hybrid algorithm', 'breast cancer'])\n", "('NONE', 'the overall accuracy offered by the employed XXXXX confirms that the effectiveness and performance of the proposed XXXXX is high', ['hybrid technique', 'hybrid system'])\n", "('method used for task', 'the numerical simulation results confirm the superiority of the proposed ainet-sl in XXXXX and XXXXX', ['solution accuracy', 'convergence speed'])\n", "('NONE', 'we consider XXXXX in leader–follower formation with the XXXXX using the potential field method', ['path planning', 'obstacle avoidance'])\n", "('method used for task', 'a novel cost-sensitive ensemble , based on XXXXX , for XXXXX', ['decision trees', 'imbalanced classification'])\n", "('method used for task', 'XXXXX are utilized to solve multi-path XXXXX', ['meta-heuristic algorithms', 'selection problem'])\n", "('method used for task', 'hybrid oc–ga method has powerful capacity in searching for more optimal XXXXX and requiring less XXXXX', ['truss structures', 'computational effort'])\n", "('method used for task', 'XXXXX according to customer-specific XXXXX , which are highly relevant to the customers’ satisfaction level', ['vehicle routing', 'time windows'])\n", "('method used for task', 'XXXXX according to customer-specific time windows , which are highly relevant to the customers’ XXXXX', ['vehicle routing', 'satisfaction level'])\n", "('method used for task', 'vehicle routing according to customer-specific XXXXX , which are highly relevant to the customers’ XXXXX', ['time windows', 'satisfaction level'])\n", "('method used for task', 'the l1/2 penalized XXXXX is able to reduce the size of the predictor even further at moderate costs for the XXXXX', ['cox model', 'prediction accuracy'])\n", "('method used for task', 'the l1/2 penalized XXXXX is suitable for XXXXX with the high dimensional biological data', ['cox model', 'survival analysis'])\n", "('NONE', 'the user-interest ontology construction is proposed by using user log profile . we describe three steps of ontology construction approach : concept selection , the generation of optimized XXXXX , the process of translating XXXXX into user-interest ontology . it is worth to be mentioned that we can revise and update the user-interest ontology using the optimized XXXXX by fca . the more time the user uses XXXXX , the richer knowledge the user-interest ontology contains', ['concept lattice', 'search engine'])\n", "('NONE', 'the user-interest XXXXX is proposed by using user log profile . we describe three steps of XXXXX approach : concept selection , the generation of optimized XXXXX , the process of translating XXXXX into user-interest ontology . it is worth to be mentioned that we can revise and update the user-interest ontology using the optimized XXXXX by fca . the more time the user uses search engine , the richer knowledge the user-interest ontology contains', ['concept lattice', 'ontology construction'])\n", "('NONE', 'the user-interest XXXXX is proposed by using user log profile . we describe three steps of XXXXX approach : concept selection , the generation of optimized XXXXX , the process of translating XXXXX into user-interest ontology . it is worth to be mentioned that we can revise and update the user-interest ontology using the optimized XXXXX by fca . the more time the user uses search engine , the richer knowledge the user-interest ontology contains', ['concept lattice', 'ontology construction'])\n", "('NONE', 'the user-interest ontology construction is proposed by using user log profile . we describe three steps of ontology construction approach : concept selection , the generation of optimized XXXXX , the process of translating XXXXX into user-interest ontology . it is worth to be mentioned that we can revise and update the user-interest ontology using the optimized XXXXX by fca . the more time the user uses XXXXX , the richer knowledge the user-interest ontology contains', ['concept lattice', 'search engine'])\n", "('NONE', 'the user-interest XXXXX is proposed by using user log profile . we describe three steps of XXXXX approach : concept selection , the generation of optimized XXXXX , the process of translating XXXXX into user-interest ontology . it is worth to be mentioned that we can revise and update the user-interest ontology using the optimized XXXXX by fca . the more time the user uses search engine , the richer knowledge the user-interest ontology contains', ['concept lattice', 'ontology construction'])\n", "('NONE', 'the user-interest XXXXX is proposed by using user log profile . we describe three steps of XXXXX approach : concept selection , the generation of optimized XXXXX , the process of translating XXXXX into user-interest ontology . it is worth to be mentioned that we can revise and update the user-interest ontology using the optimized XXXXX by fca . the more time the user uses search engine , the richer knowledge the user-interest ontology contains', ['concept lattice', 'ontology construction'])\n", "('NONE', 'the user-interest ontology construction is proposed by using user log profile . we describe three steps of ontology construction approach : concept selection , the generation of optimized XXXXX , the process of translating XXXXX into user-interest ontology . it is worth to be mentioned that we can revise and update the user-interest ontology using the optimized XXXXX by fca . the more time the user uses XXXXX , the richer knowledge the user-interest ontology contains', ['concept lattice', 'search engine'])\n", "('NONE', 'the user-interest XXXXX is proposed by using user log profile . we describe three steps of XXXXX approach : concept selection , the generation of optimized XXXXX , the process of translating XXXXX into user-interest ontology . it is worth to be mentioned that we can revise and update the user-interest ontology using the optimized XXXXX by fca . the more time the user uses search engine , the richer knowledge the user-interest ontology contains', ['concept lattice', 'ontology construction'])\n", "('NONE', 'the user-interest XXXXX is proposed by using user log profile . we describe three steps of XXXXX approach : concept selection , the generation of optimized XXXXX , the process of translating XXXXX into user-interest ontology . it is worth to be mentioned that we can revise and update the user-interest ontology using the optimized XXXXX by fca . the more time the user uses search engine , the richer knowledge the user-interest ontology contains', ['concept lattice', 'ontology construction'])\n", "('NONE', 'the user-interest XXXXX is proposed by using user log profile . we describe three steps of XXXXX approach : concept selection , the generation of optimized concept lattice , the process of translating concept lattice into user-interest ontology . it is worth to be mentioned that we can revise and update the user-interest ontology using the optimized concept lattice by fca . the more time the user uses XXXXX , the richer knowledge the user-interest ontology contains', ['search engine', 'ontology construction'])\n", "('NONE', 'the user-interest XXXXX is proposed by using user log profile . we describe three steps of XXXXX approach : concept selection , the generation of optimized concept lattice , the process of translating concept lattice into user-interest ontology . it is worth to be mentioned that we can revise and update the user-interest ontology using the optimized concept lattice by fca . the more time the user uses XXXXX , the richer knowledge the user-interest ontology contains', ['search engine', 'ontology construction'])\n", "('NONE', 'based on user-interest ontology , we propose the seed urls selection approach . the user feature concept vectors , the XXXXX pages , the base set of hits , bipartite XXXXX , and the complete bipartite XXXXX are discussed for the seed urls selection based on user-interest ontology . the advantage of the approach is combing the semantic content and link information of web pages', ['semantic web', 'directed graph'])\n", "('NONE', 'based on user-interest ontology , we propose the seed urls selection approach . the user feature concept vectors , the XXXXX pages , the base set of hits , bipartite XXXXX , and the complete bipartite XXXXX are discussed for the seed urls selection based on user-interest ontology . the advantage of the approach is combing the semantic content and link information of web pages', ['semantic web', 'directed graph'])\n", "('method used for task', 'based on user-interest ontology , we propose the seed urls selection approach . the user feature concept vectors , the XXXXX pages , the base set of hits , bipartite directed graph , and the complete bipartite directed graph are discussed for the seed urls selection based on user-interest ontology . the advantage of the approach is combing the semantic content and link information of XXXXX', ['semantic web', 'web pages'])\n", "('NONE', 'based on user-interest ontology , we propose the seed urls XXXXX . the user feature concept vectors , the XXXXX pages , the base set of hits , bipartite directed graph , and the complete bipartite directed graph are discussed for the seed urls selection based on user-interest ontology . the advantage of the approach is combing the semantic content and link information of web pages', ['semantic web', 'selection approach'])\n", "('NONE', 'based on user-interest ontology , we propose the seed urls selection approach . the user feature concept vectors , the semantic XXXXX , the base set of hits , bipartite XXXXX , and the complete bipartite XXXXX are discussed for the seed urls selection based on user-interest ontology . the advantage of the approach is combing the semantic content and link information of XXXXX', ['directed graph', 'web pages'])\n", "('NONE', 'based on user-interest ontology , we propose the seed urls XXXXX . the user feature concept vectors , the semantic web pages , the base set of hits , bipartite XXXXX , and the complete bipartite XXXXX are discussed for the seed urls selection based on user-interest ontology . the advantage of the approach is combing the semantic content and link information of web pages', ['directed graph', 'selection approach'])\n", "('method used for task', 'based on user-interest ontology , we propose the seed urls selection approach . the user feature concept vectors , the semantic web pages , the base set of hits , bipartite XXXXX , and the complete bipartite XXXXX are discussed for the seed urls selection based on user-interest ontology . the advantage of the approach is combing the semantic content and link information of web pages', ['directed graph', 'feature vectors'])\n", "('NONE', 'based on user-interest ontology , we propose the seed urls selection approach . the user feature concept vectors , the semantic XXXXX , the base set of hits , bipartite XXXXX , and the complete bipartite XXXXX are discussed for the seed urls selection based on user-interest ontology . the advantage of the approach is combing the semantic content and link information of XXXXX', ['directed graph', 'web pages'])\n", "('NONE', 'based on user-interest ontology , we propose the seed urls XXXXX . the user feature concept vectors , the semantic web pages , the base set of hits , bipartite XXXXX , and the complete bipartite XXXXX are discussed for the seed urls selection based on user-interest ontology . the advantage of the approach is combing the semantic content and link information of web pages', ['directed graph', 'selection approach'])\n", "('method used for task', 'based on user-interest ontology , we propose the seed urls selection approach . the user feature concept vectors , the semantic web pages , the base set of hits , bipartite XXXXX , and the complete bipartite XXXXX are discussed for the seed urls selection based on user-interest ontology . the advantage of the approach is combing the semantic content and link information of web pages', ['directed graph', 'feature vectors'])\n", "('NONE', 'based on user-interest ontology , we propose the seed urls XXXXX . the user feature concept vectors , the semantic XXXXX , the base set of hits , bipartite directed graph , and the complete bipartite directed graph are discussed for the seed urls selection based on user-interest ontology . the advantage of the approach is combing the semantic content and link information of XXXXX', ['web pages', 'selection approach'])\n", "('method used for task', 'based on user-interest ontology , we propose the seed urls selection approach . the user feature concept vectors , the semantic XXXXX , the base set of hits , bipartite directed graph , and the complete bipartite directed graph are discussed for the seed urls selection based on user-interest ontology . the advantage of the approach is combing the semantic content and link information of XXXXX', ['web pages', 'feature vectors'])\n", "('method used for task', 'four fhr parameters are extracted from each fpcg signal . the XXXXX and XXXXX are designed based on standard guidelines', ['membership functions', 'fuzzy rules'])\n", "('method used for task', 'geometric optimisation , XXXXX and XXXXX are combined for achieving the above purposes', ['shape grammars', 'genetic algorithms'])\n", "('NONE', 'a semi-supervised XXXXX is proposed using modified XXXXX', ['self-organizing feature map', 'change detection method'])\n", "('NONE', 'we present a novel XXXXX on evolutionary XXXXX identifying the parameters of elm', ['model based', 'membrane algorithm'])\n", "('NONE', 'the classifiers are XXXXX , XXXXX , and nearest neighbor heuristics', ['logistic regression', 'decision tree'])\n", "('NONE', 'we present the results of two XXXXX real life XXXXX in the telco industry', ['large scale', 'case studies'])\n", "('NONE', 'system based on XXXXX , XXXXX and neural network', ['support vector machine', 'genetic algorithm'])\n", "('method used for task', 'system based on XXXXX , genetic algorithm and XXXXX', ['support vector machine', 'neural network'])\n", "('method used for task', 'system based on support vector machine , XXXXX and XXXXX', ['genetic algorithm', 'neural network'])\n", "('NONE', 'a comprehensive model is proposed for quick and accurate XXXXX and identification in XXXXX', ['fault detection', 'health monitoring'])\n", "('method used for task', 'a new method which combines the XXXXX and XXXXX is developed to classify data and accurately detect and isolate faults', ['k-means clustering', 'artificial neural networks'])\n", "('method used for task', 'we associate fqfd with relative XXXXX to identify adjusted XXXXX in an fmcdm model', ['preference relation', 'criteria weights'])\n", "('NONE', 'by the adjusted XXXXX , we can avoid multiplying triangular or XXXXX , aggregating multiplied fuzzy numbers , and ranking them', ['trapezoidal fuzzy numbers', 'criteria weights'])\n", "('method used for task', 'adjusted XXXXX substitutes for original XXXXX in fmcdm through relative XXXXX , and thus adjusted XXXXX are useful to construct an fmcdm model', ['preference relation', 'criteria weights'])\n", "('method used for task', 'adjusted XXXXX substitutes for original XXXXX in fmcdm through relative XXXXX , and thus adjusted XXXXX are useful to construct an fmcdm model', ['preference relation', 'criteria weights'])\n", "('method used for task', 'adjusted XXXXX substitutes for original XXXXX in fmcdm through relative XXXXX , and thus adjusted XXXXX are useful to construct an fmcdm model', ['preference relation', 'criteria weights'])\n", "('NONE', 'a novel nonlinear XXXXX for prediction of contact area and XXXXX', ['contact pressure', 'soft computing approach'])\n", "('method used for task', 'two anfis models have been designed to correlate the XXXXX to material removal rate ( mrr ) and XXXXX ( sr )', ['process parameters', 'surface roughness'])\n", "('method used for task', 'a continuous XXXXX ( caco ) technique has been used to select the best XXXXX for maximum mrr and specified sr', ['ant colony optimization', 'process parameters'])\n", "('method used for task', 'we incorporate XXXXX to create fuzzy XXXXX for helix pairs', ['fuzzy logic', 'contact maps'])\n", "('method used for task', 'retrieved XXXXX regions are used to predict a XXXXX using bound smoothing', ['distance map', 'distance map regions'])\n", "('method used for task', 'the embed algorithm is used on the predicted XXXXX to obtain XXXXX', ['distance map', '3d structure'])\n", "('method used for task', 'an lrgf XXXXX with pole assignment technique is proposed to model the XXXXX', ['neural network', 'dynamic system'])\n", "('method used for task', 'XXXXX are constructed to fit the XXXXX with every ɛ error', ['rbf neural networks', 'singular value'])\n", "('method used for task', 'XXXXX prediction is very important for XXXXX . in this paper , an intelligent approach is adopted in order to achieve XXXXX prediction and obtain good performance', ['software engineering', 'software reliability'])\n", "('method used for task', 'XXXXX is very important for XXXXX . in this paper , an intelligent approach is adopted in order to achieve XXXXX and obtain good performance', ['software engineering', 'software reliability prediction'])\n", "('method used for task', 'XXXXX is very important for XXXXX . in this paper , an intelligent approach is adopted in order to achieve XXXXX and obtain good performance', ['software engineering', 'software reliability prediction'])\n", "('method used for task', 'XXXXX prediction is very important for software engineering . in this paper , an intelligent approach is adopted in order to achieve XXXXX prediction and obtain good performance', ['software reliability', 'software reliability prediction'])\n", "('method used for task', 'XXXXX prediction is very important for software engineering . in this paper , an intelligent approach is adopted in order to achieve XXXXX prediction and obtain good performance', ['software reliability', 'software reliability prediction'])\n", "('NONE', 'pre-processing by XXXXX can greatly reduce XXXXX , and produces improved routings when the network nodes form aggregations', ['k-means clustering', 'computation time'])\n", "('NONE', 'with a goal to enhance the scalability and accuracy of XXXXX ( rss ) , we have introduced a new model which is a combination of both XXXXX and supervised learning', ['recommendation systems', 'unsupervised learning'])\n", "('NONE', 'with a goal to enhance the scalability and accuracy of XXXXX ( rss ) , we have introduced a new model which is a combination of both unsupervised learning and XXXXX', ['recommendation systems', 'supervised learning'])\n", "('method used for task', 'with a goal to enhance the scalability and accuracy of recommendation systems ( rss ) , we have introduced a new model which is a combination of both XXXXX and XXXXX', ['unsupervised learning', 'supervised learning'])\n", "('NONE', 'the feedforward and XXXXX of the XXXXX are self-tuned based on takagi-sugeno fuzzy model', ['feedback loops', 'control algorithm'])\n", "('NONE', 'the feedforward and XXXXX of the control algorithm are self-tuned based on takagi-sugeno XXXXX', ['feedback loops', 'fuzzy model'])\n", "('method used for task', 'the feedforward and feedback loops of the XXXXX are self-tuned based on takagi-sugeno XXXXX', ['control algorithm', 'fuzzy model'])\n", "('NONE', 'fast XXXXX with XXXXX generates the trajectory of each vehicle', ['genetic algorithm', 'bezier curve'])\n", "('method used for task', 'XXXXX are trained on-line with an XXXXX based algorithm', ['neural networks', 'extended kalman filter'])\n", "('NONE', 'XXXXX of XXXXX ( fcms ) with different functions are considered', ['fixed points', 'fuzzy cognitive maps'])\n", "('method used for task', 'we give a XXXXX for sigmoidal fcms to have unique stable XXXXX', ['sufficient condition', 'fixed points'])\n", "('method used for task', 'interval type-2 XXXXX ( it2fpid ) controllers are proposed for the XXXXX ( lfc ) problem', ['load frequency control', 'fuzzy pid'])\n", "('NONE', 'for the first time in literature , the big bang–big crunch algorithm is applied to tune the XXXXX and the footprint of uncertainty ( fou ) of XXXXX', ['membership functions', 'scaling factors'])\n", "('method used for task', 'the proposed optimum it2fpid load frequency controller is superior compared to the ordinary XXXXX and conventional XXXXX', ['pid controllers', 'fuzzy pid'])\n", "('method used for task', 'hw–sw partitioning of XXXXX is formulated as a XXXXX', ['embedded system', 'multi-objective problem'])\n", "('NONE', 'various XXXXX of the problem are taken as XXXXX', ['objective functions', 'cost terms'])\n", "('method used for task', 'a XXXXX repair technique for the adaptive XXXXX is proposed', ['crossover rate', 'de algorithm'])\n", "('method used for task', 'virtual screening methods can be improved using XXXXX and XXXXX', ['neural networks', 'support vector machines'])\n", "('method used for task', 'a new XXXXX called hybrid intelligence image fusion ( hiif ) is used for fusing multimodal XXXXX', ['hybrid algorithm', 'medical images'])\n", "('NONE', 'an application of fclarans for attribute clustering and XXXXX in XXXXX has been demonstrated', ['dimensionality reduction', 'gene expression data'])\n", "('method used for task', 'XXXXX based on XXXXX and differential gene expressions are employed in the process', ['domain knowledge', 'gene ontology'])\n", "('method used for task', 'the diagnosis of XXXXX ( cvds ) is faced using a linguistic fuzzy rule-based XXXXX', ['cardiovascular diseases', 'classification system'])\n", "('NONE', 'a multiple classifier system based on neural networks/support vector machines as base classifiers , balanced subspaces to address XXXXX , a XXXXX fuser and a fuzzy diversity measure is presented', ['class imbalance', 'neural network'])\n", "('method used for task', 'a multiple classifier system based on neural networks/support vector machines as XXXXX , balanced subspaces to address XXXXX , a neural network fuser and a fuzzy diversity measure is presented', ['class imbalance', 'base classifiers'])\n", "('method used for task', 'a multiple classifier system based on neural networks/support vector machines as XXXXX , balanced subspaces to address class imbalance , a XXXXX fuser and a fuzzy diversity measure is presented', ['neural network', 'base classifiers'])\n", "('method used for task', 'experimental results show excellent XXXXX on a challenging dataset and statistical superiority compared to other XXXXX', ['classification performance', 'ensemble classifiers'])\n", "('NONE', 'analyzing the performance of 75 XXXXX on a XXXXX', ['color constancy algorithms', 'benchmark dataset'])\n", "('NONE', 'this paper solves the real world problem of tactical XXXXX , in which optimality is sought to be achieved using the evolutionary computing method of XXXXX', ['differential evolution', 'missile guidance'])\n", "('NONE', 'we have applied an improved XXXXX sde , which is improved version of basic XXXXX ( de )', ['optimization technique', 'differential evolution'])\n", "('method used for task', 'it was adapted the fuzzy probability theory to the classical XXXXX for estimating the XXXXX of a component', ['reliability analysis', 'fuzzy reliability'])\n", "('NONE', 'the multi-center XXXXX can get rid of the problems that the XXXXX is sensitive to the initial prototypes , and it can handle non-traditional curved clusters', ['initialization method', 'fcm algorithm'])\n", "('method used for task', 'lpm-fft , XXXXX , and zernike moment are utilized to extract XXXXX', ['gabor filter', 'image features'])\n", "('NONE', 'the perceptual relativity has been applied to improve the performance of classification on the sparse , noisy or XXXXX , indicating the possibility of other perceptual laws in XXXXX being considered for classification', ['imbalanced data', 'cognitive psychology'])\n", "('NONE', 'reducing the XXXXX of optimization procedure of XXXXX using weighted least squares support vector machine', ['computational time', 'gravity dams'])\n", "('NONE', 'reducing the XXXXX of optimization procedure of gravity dams using XXXXX', ['computational time', 'weighted least squares support vector machine'])\n", "('NONE', 'reducing the computational time of optimization procedure of XXXXX using XXXXX', ['gravity dams', 'weighted least squares support vector machine'])\n", "('NONE', 'finding the XXXXX of concrete XXXXX with consideration of dam–water foundation rock interaction', ['optimal shape', 'gravity dams'])\n", "('method used for task', 'a hierarchical methodology based on the same principles of conventional takagi–sugeno identification ( XXXXX and XXXXX ) , that first determines the hybrid behavior of the system and then all the other non-linearities', ['principal components', 'fuzzy clustering'])\n", "('NONE', 'different XXXXX are compared and the impacts of parallel parameter on the proposed XXXXX are discussed', ['chaotic maps', 'hybrid algorithm'])\n", "('NONE', 'this paper proposes a self-organizing XXXXX as a XXXXX for ovarian cancer diagnoses', ['neural fuzzy system', 'decision support system'])\n", "('method used for task', 'XXXXX and XXXXX are performed during training', ['feature selection', 'attribute reduction'])\n", "('method used for task', 'we consider XXXXX and XXXXX to represent ambiguous , uncertain or imprecise information', ['fuzzy logic', 'fuzzy sets'])\n", "('method used for task', 'we consider XXXXX and fuzzy sets to represent ambiguous , uncertain or XXXXX', ['fuzzy logic', 'imprecise information'])\n", "('method used for task', 'we consider fuzzy logic and XXXXX to represent ambiguous , uncertain or XXXXX', ['fuzzy sets', 'imprecise information'])\n", "('method used for task', 'the integrated hybrid methodology is XXXXX and XXXXX for risk assessment', ['fuzzy ahp', 'fuzzy topsis'])\n", "('method used for task', 'the integrated hybrid methodology is XXXXX and fuzzy topsis for XXXXX', ['fuzzy ahp', 'risk assessment'])\n", "('method used for task', 'the integrated hybrid methodology is fuzzy ahp and XXXXX for XXXXX', ['fuzzy topsis', 'risk assessment'])\n", "('NONE', 'embeds XXXXX using a parameterized function that is automatically estimated using XXXXX', ['domain knowledge', 'stochastic optimization'])\n", "('method used for task', 'we propose a new type-2 XXXXX ( fs ) learned through structure and XXXXX', ['neural fuzzy system', 'parameter learning'])\n", "('method used for task', 'XXXXX and XXXXX are applied to markerless motion capture', ['evolutionary algorithms', 'particle filters'])\n", "('method used for task', 'XXXXX and particle filters are applied to markerless XXXXX', ['evolutionary algorithms', 'motion capture'])\n", "('method used for task', 'evolutionary algorithms and XXXXX are applied to markerless XXXXX', ['particle filters', 'motion capture'])\n", "('NONE', 'non-parametric tests show that XXXXX outperform XXXXX', ['evolutionary algorithms', 'particle filters'])\n", "('NONE', 'XXXXX and hierarchical strategies deal better with XXXXX', ['evolutionary algorithms', 'high dimensionality'])\n", "('method used for task', 'we present our versions of two well-known XXXXX ( nsgaii and spea2 ) applied to this XXXXX', ['multiobjective evolutionary algorithms', 'mobile network problem'])\n", "('method used for task', 'feature-level and XXXXX is explored using svm and rf classifiers , wrapper-based amde XXXXX is also investigated', ['decision-level fusion', 'feature selection'])\n", "('method used for task', 'feature-level and XXXXX is explored using svm and XXXXX , wrapper-based amde feature selection is also investigated', ['decision-level fusion', 'rf classifiers'])\n", "('NONE', 'feature-level and decision-level fusion is explored using svm and XXXXX , wrapper-based amde XXXXX is also investigated', ['feature selection', 'rf classifiers'])\n", "('method used for task', 'we contribute to XXXXX decomposition approaches by introducing a simple feature oriented technique for ensemble design , which is based on partitioning the XXXXX by XXXXX', ['feature set', 'k-means clustering'])\n", "('NONE', 'results indicate that ensemble of XXXXX , induced using the proposed technique , significantly improve decision-level and outperform XXXXX', ['feature-level fusion', 'rf classifiers'])\n", "('NONE', 'a set of three evolving XXXXX are derived by an online XXXXX', ['tsk fuzzy models', 'identification algorithm'])\n", "('method used for task', 'an improved fkp-mco classifier based on XXXXX , kernel technique , and XXXXX is proposed and is used for predicting protein–protein interaction hot spots', ['fuzzification method', 'penalty factors'])\n", "('NONE', 'the XXXXX uses XXXXX and moving average technical index to predict stock price trends', ['hybrid model', 'linear model'])\n", "('NONE', 'a XXXXX is proposed for selection of abstract data types implementations during the execution of a XXXXX', ['support vector machine', 'software system'])\n", "('NONE', 'the considered XXXXX confirm a good performance of the proposed model and show that using XXXXX for the data selection problem is worth being further investigated', ['computational experiments', 'machine learning techniques'])\n", "('method used for task', 'this study proposes a XXXXX based on the fuzzy decision-making trail and evaluation laboratory ( dematel ) and fuzzy multi-criteria decision-making ( fmcdm ) for XXXXX in sc', ['prediction framework', 'km adoption'])\n", "('NONE', 'it also enables organizations to decide whether to initiate XXXXX , restrain adoption or undertake remedial improvements to increase the possibility of successful XXXXX in sc', ['knowledge management', 'km adoption'])\n", "('NONE', 'XXXXX acts as XXXXX for ga', ['support vector machine', 'objective function'])\n", "('NONE', 'mlp is considered for each XXXXX for accurate estimation of dry XXXXX', ['soil types', 'unit weights'])\n", "('method used for task', 'a new pso-based XXXXX is presented to solve dynamic economic dispatch problems with XXXXX', ['hybrid algorithm', 'valve-point effects'])\n", "('method used for task', 'we propose a novel XXXXX for applicability of hyper-heuristic techniques on XXXXX', ['hybrid strategy', 'dynamic environments'])\n", "('method used for task', 'performance of our method is validated with the dynamic XXXXX and the moving XXXXX', ['generalized assignment problem', 'peaks benchmark'])\n", "('NONE', 'point out the drawbacks of the XXXXX and the parameter settings of XXXXX ( de )', ['crossover operator', 'differential evolution'])\n", "('method used for task', 'we propose a fast XXXXX to uncover XXXXX in networks', ['memetic algorithm', 'community structure'])\n", "('method used for task', 'to the best of our knowledge this is the first attempt to apply the XXXXX to a real-industrial XXXXX', ['cuckoo search algorithm', 'scheduling problem'])\n", "('method used for task', 'this paper proposes a new XXXXX for solving XXXXX', ['heuristic algorithm', 'optimization problems'])\n", "('method used for task', 'this XXXXX is inspired of trading the shares on XXXXX', ['optimization algorithm', 'stock market'])\n", "('NONE', 'pitfalls in the comparison of XXXXX within the field of applied XXXXX are quite common', ['computational experiments', 'evolutionary computing'])\n", "('method used for task', 'common pitfalls found in replication and comparison of XXXXX for economic XXXXX problem are presented', ['computational experiments', 'load dispatch'])\n", "('method used for task', 'guidelines on setting and conducting XXXXX for XXXXX are provided', ['computational experiments', 'evolutionary algorithms'])\n", "('NONE', 'XXXXX with a set of large-scale instances show that the mdsfl can be an efficient alternative for solving tightly constrained 01 XXXXX', ['computational experiments', 'knapsack problems'])\n", "('method used for task', 'the partition based clustering algorithms k-means and XXXXX algorithms are taken for analysis via its XXXXX', ['fuzzy c-means', 'computational time'])\n", "('method used for task', 'the distribution of data points by XXXXX is even to all the XXXXX , but , it is not even by the fcm algorithm', ['k-means algorithm', 'data centers'])\n", "('NONE', 'the distribution of XXXXX by XXXXX is even to all the data centers , but , it is not even by the fcm algorithm', ['k-means algorithm', 'data points'])\n", "('method used for task', 'the distribution of data points by XXXXX is even to all the data centers , but , it is not even by the XXXXX', ['k-means algorithm', 'fcm algorithm'])\n", "('method used for task', 'the distribution of XXXXX by k-means algorithm is even to all the XXXXX , but , it is not even by the fcm algorithm', ['data centers', 'data points'])\n", "('method used for task', 'the distribution of data points by k-means algorithm is even to all the XXXXX , but , it is not even by the XXXXX', ['data centers', 'fcm algorithm'])\n", "('method used for task', 'the distribution of XXXXX by k-means algorithm is even to all the data centers , but , it is not even by the XXXXX', ['data points', 'fcm algorithm'])\n", "('NONE', 'from the XXXXX , the XXXXX of k-means algorithm is less than the fcm algorithm', ['experimental analysis', 'computational time'])\n", "('NONE', 'from the XXXXX , the computational time of XXXXX is less than the fcm algorithm', ['experimental analysis', 'k-means algorithm'])\n", "('NONE', 'from the experimental analysis , the XXXXX of XXXXX is less than the fcm algorithm', ['computational time', 'k-means algorithm'])\n", "('NONE', 'an elitist self-adaptive step-size search ( esass ) algorithm is proposed for XXXXX of XXXXX subject to stress and displacement constraints', ['optimum design', 'truss structures'])\n", "('NONE', 'the XXXXX of the technique is accelerated through avoiding unnecessary analyses throughout the XXXXX using the so-called upper bound strategy ( ubs )', ['computational efficiency', 'optimization process'])\n", "('NONE', 'the esass algorithm is capable of locating reasonable solutions to optimum sizing problems of XXXXX with considerably less XXXXX', ['truss structures', 'computational effort'])\n", "('method used for task', 'XXXXX probability is adapted based on effectiveness of XXXXX operator', ['local search', 'local search operator'])\n", "('method used for task', 'maximum profit has been compared by XXXXX and XXXXX based XXXXX', ['genetic algorithm', 'fuzzy simulation'])\n", "('method used for task', 'maximum profit has been compared by XXXXX and XXXXX based XXXXX', ['fuzzy simulation', 'genetic algorithm'])\n", "('method used for task', 'the concept of the XXXXX is used to collect the real-time XXXXX and derives the appropriate paths for users', ['cellular automata', 'road conditions'])\n", "('NONE', 'the proposed work makes use of the XXXXX and the concept of the XXXXX to reduce the computational complexity', ['hierarchical structure', 'cellular automata'])\n", "('method used for task', 'the proposed work makes use of the XXXXX and the concept of the cellular automata to reduce the XXXXX', ['hierarchical structure', 'computational complexity'])\n", "('method used for task', 'the proposed work makes use of the hierarchical structure and the concept of the XXXXX to reduce the XXXXX', ['cellular automata', 'computational complexity'])\n", "('NONE', 'fuzzy XXXXX ( fsd ) intertwines XXXXX and XXXXX', ['statistical downscaling', 'fuzzy system'])\n", "('NONE', 'fuzzy XXXXX ( fsd ) intertwines XXXXX and XXXXX', ['fuzzy system', 'statistical downscaling'])\n", "('NONE', 'the XXXXX can be executed rapidly or with XXXXX by even systems based on fixed-point processors', ['control algorithm', 'low power consumption'])\n", "('method used for task', 'develops five XXXXX for parallel-machine and inbound-trucks sequencing in multi door XXXXX', ['hybrid metaheuristics', 'cross docking'])\n", "('method used for task', 'the hybrid simulated-annealing tabu-search algorithm is the best if XXXXX is used the XXXXX instead', ['cpu time', 'stopping criterion'])\n", "('NONE', 'aim at preventing overfitting of the state of the XXXXX ( XXXXX ) and improves upon other ( indictive method )', ['least squares', 'art methods'])\n", "('method used for task', 'adaptive XXXXX have been constructed in the proposed technique for both bell-shaped and triangle-shaped XXXXX', ['fuzzy membership functions', 'fuzzy sets'])\n", "('NONE', 'idea of multiple XXXXX has been incorporated in an effective manner for the removal of XXXXX from degraded images', ['fuzzy membership functions', 'impulse noise'])\n", "('method used for task', 'detailed XXXXX is presented , showing that different XXXXX perform better for different types of images at different noise ratio', ['sensitivity analysis', 'fuzzy membership functions'])\n", "('method used for task', 'detailed XXXXX is presented , showing that different fuzzy membership functions perform better for different types of images at different XXXXX', ['sensitivity analysis', 'noise ratio'])\n", "('NONE', 'detailed sensitivity analysis is presented , showing that different XXXXX perform better for different types of images at different XXXXX', ['fuzzy membership functions', 'noise ratio'])\n", "('method used for task', 'performance results based on the igd metric show that the XXXXX is overall superior to versions with single XXXXX', ['hybrid algorithm', 'search operators'])\n", "('method used for task', 'the proposed method is experimented on 13 unconstrained XXXXX and 24 constrained XXXXX , 20 XXXXX and 22 real life problems from the literature', ['benchmark problems', 'mechanical design problems'])\n", "('method used for task', 'the proposed method is experimented on 13 unconstrained XXXXX and 24 constrained XXXXX , 20 XXXXX and 22 real life problems from the literature', ['benchmark problems', 'mechanical design problems'])\n", "('method used for task', 'describes an XXXXX and a multi-round XXXXX for solving the problem', ['heuristic algorithm', 'integer programming formulation'])\n", "('NONE', 'the paper contributes to support the hypothesis that hybrid and XXXXX can be used in XXXXX without necessarily incurring significantly higher distance-based costs', ['electric vehicles', 'routing problems'])\n", "('NONE', 'compares XXXXX and fuzzy ahp methods concerning the problem of XXXXX based on a set of seven criteria', ['fuzzy topsis', 'supplier selection'])\n", "('NONE', 'further work can explore alternative approaches to avoid nulling weights of the criteria and XXXXX in XXXXX', ['rank reversal', 'fuzzy ahp'])\n", "('NONE', 'an adaptive hybrid control system using rrsefnn for pmsm servo drive is proposed.the computed torque control is designed to stabilize the pmsm servo drive.the rrsefnn is used to estimate the nonlinear lumped uncertainty of the pmsm drive.the online adaptive laws are derived using the XXXXX analysis.all XXXXX are implemented using tms320c31 dsp-based control computer', ['lyapunov stability', 'control algorithms'])\n", "('method used for task', 'an adaptive hybrid control system using rrsefnn for pmsm servo drive is proposed.the computed XXXXX is designed to stabilize the pmsm servo drive.the rrsefnn is used to estimate the nonlinear lumped uncertainty of the pmsm drive.the online adaptive laws are derived using the XXXXX analysis.all control algorithms are implemented using tms320c31 dsp-based control computer', ['lyapunov stability', 'torque control'])\n", "('NONE', 'an adaptive hybrid control system using rrsefnn for pmsm servo drive is proposed.the computed torque control is designed to stabilize the pmsm servo drive.the rrsefnn is used to estimate the nonlinear lumped uncertainty of the pmsm drive.the online adaptive laws are derived using the XXXXX analysis.all control algorithms are implemented using tms320c31 dsp-based XXXXX', ['lyapunov stability', 'control computer'])\n", "('method used for task', 'an adaptive hybrid control system using rrsefnn for pmsm servo drive is proposed.the computed XXXXX is designed to stabilize the pmsm servo drive.the rrsefnn is used to estimate the nonlinear lumped uncertainty of the pmsm drive.the online adaptive laws are derived using the lyapunov stability analysis.all XXXXX are implemented using tms320c31 dsp-based control computer', ['control algorithms', 'torque control'])\n", "('NONE', 'an adaptive hybrid control system using rrsefnn for pmsm servo drive is proposed.the computed torque control is designed to stabilize the pmsm servo drive.the rrsefnn is used to estimate the nonlinear lumped uncertainty of the pmsm drive.the online adaptive laws are derived using the lyapunov stability analysis.all XXXXX are implemented using tms320c31 dsp-based XXXXX', ['control algorithms', 'control computer'])\n", "('method used for task', 'an adaptive hybrid control system using rrsefnn for pmsm servo drive is proposed.the computed XXXXX is designed to stabilize the pmsm servo drive.the rrsefnn is used to estimate the nonlinear lumped uncertainty of the pmsm drive.the online adaptive laws are derived using the lyapunov stability analysis.all control algorithms are implemented using tms320c31 dsp-based XXXXX', ['torque control', 'control computer'])\n", "('NONE', 'proposed system used ensemble XXXXX with svm as XXXXX for the classification of XXXXX as normal or tumor', ['brain images', 'base classifier'])\n", "('NONE', 'proposed system used ensemble XXXXX with svm as XXXXX for the classification of XXXXX as normal or tumor', ['brain images', 'base classifier'])\n", "('method used for task', 'XXXXX based weighted majority voting scheme has been used for XXXXX', ['genetic algorithm', 'ensemble classifiers'])\n", "('NONE', 'the problem of finding the expected XXXXX in XXXXX has numerous applications', ['shortest path', 'stochastic networks'])\n", "('NONE', 'the proposed scheme is subjected to some XXXXX as well as XXXXX suites', ['security analysis', 'statistical tests'])\n", "('method used for task', 'we propose a XXXXX for XXXXX', ['human resources', 'performance evaluation method'])\n", "('NONE', 'the result shows that the proposed algorithm can generate better XXXXX within shorter XXXXX and stable convergence characteristics compared to fapso , pso and ga', ['computational time', 'quality solution'])\n", "('method used for task', 'the result shows that the proposed algorithm can generate better quality solution within shorter XXXXX and stable XXXXX compared to fapso , pso and ga', ['computational time', 'convergence characteristics'])\n", "('method used for task', 'the result shows that the proposed algorithm can generate better XXXXX within shorter computational time and stable XXXXX compared to fapso , pso and ga', ['quality solution', 'convergence characteristics'])\n", "('method used for task', 'existing XXXXX detection techniques are classifying the given signal as normal voice and XXXXX . but no technique is to detect the type of pathology ( analyzed the techniques published during last five years )', ['pathological voice', 'voice pathology'])\n", "('method used for task', 'a two level system has been proposed for detecting the type of XXXXX . in the first level , the system identifies whether the given voice is normal or not . if it is recognized as XXXXX , in the second level which aims at identifying the particular type of pathology', ['pathological voice', 'voice pathology'])\n", "('NONE', 'a two-level model is proposed for real-time XXXXX in a XXXXX', ['path planning', 'dynamic environment'])\n", "('NONE', 'in case of XXXXX the lack of XXXXX of small magnitude is only occasionally observed', ['benchmark problems', 'difference vectors'])\n", "('method used for task', 'the algorithm is tested using XXXXX and a XXXXX', ['benchmark functions', 'real-world application'])\n", "('method used for task', 'XXXXX was used to generate the XXXXX', ['subtractive clustering', 'membership functions'])\n", "('NONE', 'XXXXX allowed training the anfis with XXXXX with noise', ['subtractive clustering', 'experimental data'])\n", "('NONE', 'results illustrate that the fuzzy calculator is an efficient tool to propagate XXXXX regarding non-interactive as well as interactive fuzzy XXXXX', ['epistemic uncertainty', 'input variables'])\n", "('method used for task', 'in this type of strategy to combine the potential for aggregation of information/knowledge of the XXXXX for processing XXXXX in the bayesian network simplicity we will call this combination of fuzzy/bayesian network', ['fuzzy sets theory', 'input uncertainties'])\n", "('NONE', 'to illustrate the efficiency of the proposed methodology , the problem of XXXXX in the stator winding of XXXXX is presented', ['fault detection', 'induction machine'])\n", "('method used for task', 'the new strategy is based on the potential for aggregation of information/knowledge the XXXXX for processing input uncertainties in the XXXXX', ['fuzzy sets theory', 'bayesian network'])\n", "('method used for task', 'the new strategy is based on the potential for aggregation of information/knowledge the XXXXX for processing XXXXX in the bayesian network', ['fuzzy sets theory', 'input uncertainties'])\n", "('NONE', 'the new strategy is based on the potential for aggregation of information/knowledge the fuzzy sets theory for processing XXXXX in the XXXXX', ['bayesian network', 'input uncertainties'])\n", "('NONE', 'we propose a new XXXXX for segmentation of XXXXX', ['hybrid model', 'sar images'])\n", "('NONE', 'XXXXX captures the textural information of XXXXX more precisely', ['neutrosophic set', 'sar image'])\n", "('NONE', 'the proposed method can segment the XXXXX containing XXXXX', ['sar images', 'speckle noise'])\n", "('method used for task', 'the XXXXX is formulated as a XXXXX', ['multi-objective optimization problem', 'optimal power flow problem'])\n", "('method used for task', 'modifying the XXXXX by applying opposition based learning ( obl ) to enhance its XXXXX', ['tlbo algorithm', 'search ability'])\n", "('NONE', 'we propose XXXXX based XXXXX to train the proposed anns', ['genetic algorithm', 'learning algorithm'])\n", "('method used for task', 'fuzzy analytical XXXXX is efficient to extract customers’ preferences for XXXXX', ['hierarchy process', 'core attributes'])\n", "('NONE', 'we propose a set of XXXXX to formally reduce the problem of XXXXX pctlc to the problem of XXXXX pctl', ['reduction rules', 'model checking'])\n", "('method used for task', 'the application of XXXXX and biogeography based optimization to generate near-optimal XXXXX is being proposed', ['genetic algorithm', 'golomb ruler sequences'])\n", "('NONE', 'both the approaches produce near-optimal XXXXX very efficiently in reasonable XXXXX', ['execution time', 'golomb ruler sequences'])\n", "('NONE', 'in particular , there have been recent applications of XXXXX in the fields of XXXXX , classification and clustering , where it has helped improving results over type-1 fuzzy logic', ['type-2 fuzzy logic', 'pattern recognition'])\n", "('method used for task', 'since sample data may include uncertainties coming from measurement systems and environmental conditions , XXXXX and/or XXXXX can be used to capture these uncertainties', ['fuzzy numbers', 'linguistic variables'])\n", "('method used for task', 'since XXXXX may include uncertainties coming from measurement systems and environmental conditions , XXXXX and/or linguistic variables can be used to capture these uncertainties', ['fuzzy numbers', 'sample data'])\n", "('method used for task', 'since XXXXX may include uncertainties coming from measurement systems and environmental conditions , fuzzy numbers and/or XXXXX can be used to capture these uncertainties', ['linguistic variables', 'sample data'])\n", "('NONE', 'in this paper , one of the most popular XXXXX , exponentially weighted moving average XXXXX ( ewma ) for univariate data are developed under fuzzy environment', ['control charts', 'control chart'])\n", "('method used for task', 'in this paper , one of the most popular XXXXX , exponentially weighted moving average control chart ( ewma ) for univariate data are developed under XXXXX', ['control charts', 'fuzzy environment'])\n", "('method used for task', 'in this paper , one of the most popular control charts , exponentially weighted moving average XXXXX ( ewma ) for univariate data are developed under XXXXX', ['control chart', 'fuzzy environment'])\n", "('NONE', 'automated retinal vessel segmentation technique using XXXXX for screening of XXXXX', ['neural network', 'diabetic retinopathy'])\n", "('NONE', 'proposed method has XXXXX of the vasculature in XXXXX', ['automatic recognition', 'retinal images'])\n", "('NONE', 'we compare som and ghsom models based on the XXXXX and goodness of XXXXX to find the efficacy of the performance on a set of 15 benchmarked problems', ['network architecture', 'cell formation'])\n", "('NONE', 'i propose a growing XXXXX developed from perspective of a XXXXX', ['neural network', 'generative model'])\n", "('NONE', 'surveying various type-2 fuzzy disciplines including XXXXX , XXXXX , etc', ['fuzzy systems', 'fuzzy clustering'])\n", "('method used for task', 'this paper presents an effective biased random-key XXXXX for the tactical XXXXX', ['genetic algorithm', 'berth allocation problem'])\n", "('method used for task', 'correlation between grasping XXXXX and embedded XXXXX', ['objects weight', 'sensor stress'])\n", "('method used for task', 'XXXXX is performed using two XXXXX', ['parametric identification', 'soft computing techniques'])\n", "('method used for task', 'good agreement is found between XXXXX and XXXXX', ['experimental data', 'numerical simulations'])\n", "('method used for task', 'new email spam detection model based on XXXXX and XXXXX ( nsa–pso ) is implemented', ['negative selection algorithm', 'particle swarm optimization'])\n", "('NONE', 'random detector generation is replaced with XXXXX ; XXXXX and threshold value were studied to select distinctive features for spam detection', ['particle swarm optimization', 'distance measure'])\n", "('NONE', 'this paper proposes a multi-objective XXXXX of type-2 XXXXX', ['genetic optimization', 'fuzzy controllers'])\n", "('method used for task', 'the adaptive relevance vector machine is compared to the XXXXX and the XXXXX in the estimation', ['artificial neural networks', 'support vector machine'])\n", "('method used for task', 'the adaptive XXXXX machine is compared to the XXXXX and the support vector machine in the estimation', ['artificial neural networks', 'relevance vector'])\n", "('method used for task', 'the adaptive XXXXX machine is compared to the artificial neural networks and the XXXXX in the estimation', ['support vector machine', 'relevance vector'])\n", "('NONE', 'a transformation mechanism including three XXXXX between intuitionistic XXXXX and intuitionistic multiplicative preference relations are established', ['fuzzy preference relations', 'transformation functions'])\n", "('NONE', 'a new multi-objective method based on thd , htll , mllf , and total apfs currents to eliminate XXXXX in the electrical XXXXX is taken into account', ['harmonic distortion', 'distribution network'])\n", "('method used for task', 'two XXXXX from literature are studied and the results illustrate significant improvement in structural weight compared to those of the conventional XXXXX', ['numerical examples', 'design methods'])\n", "('NONE', 'a novel XXXXX of hybrid neuro-fuzzy for XXXXX of four area power system is presented', ['load frequency control', 'control approach'])\n", "('method used for task', 'XXXXX is carried out by using fuzzy , ann , anfis and conventional pi and XXXXX approaches', ['performance evaluation', 'pid control'])\n", "('NONE', 'single-chip embedded mimo autonomous XXXXX with on-line learning capability : small footprint , XXXXX , transparent device', ['intelligent agent', 'low power'])\n", "('NONE', 'we know the XXXXX of the XXXXX affecting a component', ['stochastic model', 'degradation process'])\n", "('NONE', 'we develop a method based on XXXXX of evidence and XXXXX to propagate the uncertainty', ['dempster–shafer theory', 'fuzzy random variables'])\n", "('method used for task', 'only XXXXX is applied , but it can reach a high security and less XXXXX', ['diffusion function', 'time cost'])\n", "('NONE', 'we examined the XXXXX on sunspots , XXXXX and indian stock data', ['model accuracy', 'electricity price'])\n", "('method used for task', 'binary XXXXX ( bgsa ) is proposed to solve XXXXX', ['gravitational search algorithm', 'unit commitment'])\n", "('NONE', 'we developed an XXXXX ( ann ) model to mimic route choice behaviour in crowds which achieved a XXXXX of 86 %', ['artificial neural network', 'prediction accuracy'])\n", "('method used for task', 'a XXXXX is proposed to make a XXXXX , called as physarum algorithm , more efficient', ['heuristic method', 'bio-inspired algorithm'])\n", "('method used for task', 'XXXXX is employed for the determination of XXXXX', ['pso algorithm', 'model parameters'])\n", "('NONE', \"an orthogonal forward selection algorithm is proposed for constructing radial basis function classifiers based on maximises the leave-one-out XXXXX between the classifier 's predicted XXXXX and the true XXXXX\", ['mutual information', 'class labels'])\n", "('NONE', \"an orthogonal forward selection algorithm is proposed for constructing radial basis function classifiers based on maximises the leave-one-out XXXXX between the classifier 's predicted XXXXX and the true XXXXX\", ['mutual information', 'class labels'])\n", "('method used for task', 'a novel XXXXX for XXXXX ( ba ) has been proposed', ['bees algorithm', 'initialization algorithm'])\n", "('NONE', 'a dead-zone XXXXX can resolve the poblem of over-trained XXXXX', ['neural network', 'parameter modification'])\n", "('NONE', 'algorithms used in the study are XXXXX ( es ) , XXXXX ( ga ) , differential evolution ( de ) , adaptive differential evolution ( jade ) , particle swarm optimization ( pso ) , artificial bee colony ( abc ) , cuckoo search ( cs ) and differential search algorithm ( ds )', ['evolution strategy', 'genetic algorithm'])\n", "('NONE', 'algorithms used in the study are XXXXX ( es ) , genetic algorithm ( ga ) , differential evolution ( de ) , adaptive differential evolution ( jade ) , XXXXX ( pso ) , artificial bee colony ( abc ) , cuckoo search ( cs ) and differential search algorithm ( ds )', ['evolution strategy', 'particle swarm optimization'])\n", "('method used for task', 'algorithms used in the study are XXXXX ( es ) , genetic algorithm ( ga ) , differential evolution ( de ) , adaptive differential evolution ( jade ) , particle swarm optimization ( pso ) , artificial bee colony ( abc ) , XXXXX ( cs ) and differential search algorithm ( ds )', ['evolution strategy', 'cuckoo search'])\n", "('method used for task', 'algorithms used in the study are XXXXX ( es ) , genetic algorithm ( ga ) , differential evolution ( de ) , adaptive differential evolution ( jade ) , particle swarm optimization ( pso ) , artificial bee colony ( abc ) , cuckoo search ( cs ) and differential XXXXX ( ds )', ['evolution strategy', 'search algorithm'])\n", "('NONE', 'algorithms used in the study are evolution strategy ( es ) , XXXXX ( ga ) , differential evolution ( de ) , adaptive differential evolution ( jade ) , XXXXX ( pso ) , artificial bee colony ( abc ) , cuckoo search ( cs ) and differential search algorithm ( ds )', ['genetic algorithm', 'particle swarm optimization'])\n", "('method used for task', 'algorithms used in the study are evolution strategy ( es ) , XXXXX ( ga ) , differential evolution ( de ) , adaptive differential evolution ( jade ) , particle swarm optimization ( pso ) , artificial bee colony ( abc ) , XXXXX ( cs ) and differential search algorithm ( ds )', ['genetic algorithm', 'cuckoo search'])\n", "('method used for task', 'algorithms used in the study are evolution strategy ( es ) , XXXXX ( ga ) , differential evolution ( de ) , adaptive differential evolution ( jade ) , particle swarm optimization ( pso ) , artificial bee colony ( abc ) , cuckoo search ( cs ) and differential XXXXX ( ds )', ['genetic algorithm', 'search algorithm'])\n", "('NONE', 'algorithms used in the study are evolution strategy ( es ) , genetic algorithm ( ga ) , differential evolution ( de ) , adaptive differential evolution ( jade ) , XXXXX ( pso ) , artificial bee colony ( abc ) , XXXXX ( cs ) and differential search algorithm ( ds )', ['particle swarm optimization', 'cuckoo search'])\n", "('method used for task', 'algorithms used in the study are evolution strategy ( es ) , genetic algorithm ( ga ) , differential evolution ( de ) , adaptive differential evolution ( jade ) , XXXXX ( pso ) , artificial bee colony ( abc ) , cuckoo search ( cs ) and differential XXXXX ( ds )', ['particle swarm optimization', 'search algorithm'])\n", "('method used for task', 'algorithms used in the study are evolution strategy ( es ) , genetic algorithm ( ga ) , differential evolution ( de ) , adaptive differential evolution ( jade ) , particle swarm optimization ( pso ) , artificial bee colony ( abc ) , XXXXX ( cs ) and differential XXXXX ( ds )', ['cuckoo search', 'search algorithm'])\n", "('method used for task', 'the artificial bee colony algorithm and differential XXXXX are the most XXXXX', ['search algorithm', 'robust algorithms'])\n", "('method used for task', 'the artificial bee XXXXX and differential XXXXX are the most robust algorithms', ['search algorithm', 'colony algorithm'])\n", "('method used for task', 'the artificial bee XXXXX and differential search algorithm are the most XXXXX', ['robust algorithms', 'colony algorithm'])\n", "('method used for task', 'comprehensive features ( cf ) include XXXXX ( sa ) , technical analysis ( ta ) and trend-based XXXXX ( tbsm )', ['sentiment analysis', 'segmentation method'])\n", "('method used for task', 'XXXXX of each dimension is divided into many equal subregions . according to XXXXX in particles’ historical best position , some dominant subregions can be determined', ['search space', 'statistics information'])\n", "('method used for task', 'XXXXX for XXXXX is considered', ['fuzzy demand', 'location problem'])\n", "('NONE', 'the efficiency and the effectiveness of the XXXXX in examined using XXXXX', ['hybrid algorithm', 'numerical examples'])\n", "('NONE', 'XXXXX are one of the most common choices reported in the literature for tuning XXXXX', ['evolutionary algorithms', 'fuzzy logic controllers'])\n", "('method used for task', 'an alternative is the simple XXXXX , which is a methodology designed to improve the response of type-1 XXXXX', ['fuzzy logic controllers', 'tuning algorithm'])\n", "('method used for task', 'an extension of the simple XXXXX for type-2 XXXXX is presented', ['fuzzy controllers', 'tuning algorithm'])\n", "('NONE', 'to incorporate XXXXX a novel XXXXX is defined', ['contextual information', 'energy function'])\n", "('NONE', 'the XXXXX of XXXXX for jusras problem grows exponentially with the number of users and receives antennas', ['computational complexity', 'exhaustive search'])\n", "('method used for task', 'we apply XXXXX ( bpso ) to the joint XXXXX and receive antenna selection problem', ['binary particle swarm optimization', 'user scheduling'])\n", "('NONE', 'in addition to applying the conventional bpso to jusras , we also present a specific improvement to this population-based XXXXX ; namely , we feed cyclically shifted XXXXX , so that the number of iterations until reaching an acceptable solution is reduced', ['heuristic algorithm', 'initial population'])\n", "('method used for task', 'a multi-agent based XXXXX is proposed for XXXXX', ['optimization approach', 'single machine scheduling problem'])\n", "('method used for task', 'an approach based on evolutionary and XXXXX is proposed for solving the XXXXX in crop growth dynamic models', ['bio-inspired algorithms', 'parameter estimation problem'])\n", "('NONE', 'XXXXX showed the best performance in solving the XXXXX of a dynamic crop growth model', ['differential evolution algorithm', 'parameter estimation problem'])\n", "('method used for task', 'a XXXXX and anova were applied to quantitatively evaluate the efficiency and effectiveness of XXXXX , cma-es , pso , abc and lse algorithms', ['statistical analysis', 'differential evolution'])\n", "('method used for task', 'this work presents flmicmac , an improved version of micmac based on computing with words , XXXXX and XXXXX', ['fuzzy numbers', 'linguistic variables'])\n", "('NONE', 'a new method is proposed for the exact analytical XXXXX of decomposable ts XXXXX with singleton and linear consequents', ['fuzzy systems', 'inverse mapping'])\n", "('NONE', 'the proposed method simplifies the inversion of decomposable ts XXXXX having high number of XXXXX', ['fuzzy systems', 'input variables'])\n", "('method used for task', 'we discuss a general framework of XXXXX for XXXXX and time series analysis', ['computational intelligence', 'signal processing'])\n", "('method used for task', 'we discuss a general framework of XXXXX for signal processing and XXXXX', ['computational intelligence', 'time series analysis'])\n", "('method used for task', 'we discuss a general framework of computational intelligence for XXXXX and XXXXX', ['signal processing', 'time series analysis'])\n", "('method used for task', 'multi-criteria decision making model is established for XXXXX and XXXXX', ['supplier evaluation', 'selection problem'])\n", "('method used for task', 'XXXXX ( nn ) for fuzzy multi criteria decision making and adaptive neuro-fuzzy XXXXX ( anfis ) models are compared for multi-criteria decision making in supplier evaluation and selection problem solution', ['neural networks', 'inference system'])\n", "('method used for task', 'XXXXX ( nn ) for fuzzy multi criteria decision making and adaptive neuro-fuzzy inference system ( anfis ) models are compared for XXXXX in supplier evaluation and selection problem solution', ['neural networks', 'multi-criteria decision making'])\n", "('method used for task', 'XXXXX ( nn ) for fuzzy multi criteria decision making and adaptive neuro-fuzzy inference system ( anfis ) models are compared for multi-criteria decision making in XXXXX and selection problem solution', ['neural networks', 'supplier evaluation'])\n", "('method used for task', 'neural networks ( nn ) for fuzzy multi criteria decision making and adaptive neuro-fuzzy XXXXX ( anfis ) models are compared for XXXXX in supplier evaluation and selection problem solution', ['inference system', 'multi-criteria decision making'])\n", "('method used for task', 'neural networks ( nn ) for fuzzy multi criteria decision making and adaptive neuro-fuzzy XXXXX ( anfis ) models are compared for multi-criteria decision making in XXXXX and selection problem solution', ['inference system', 'supplier evaluation'])\n", "('NONE', 'neural networks ( nn ) for fuzzy multi criteria decision making and adaptive neuro-fuzzy inference system ( anfis ) models are compared for XXXXX in XXXXX and selection problem solution', ['multi-criteria decision making', 'supplier evaluation'])\n", "('NONE', 'XXXXX of the XXXXX have found with ahp', ['performance criteria', 'importance weights'])\n", "('method used for task', 'the proposed approach is a combination of XXXXX ( pso ) and XXXXX ( lfpso )', ['particle swarm optimization', 'levy flight'])\n", "('NONE', 'the performance and accuracy of the lfpso are examined on numerical XXXXX especially XXXXX', ['benchmark functions', 'multimodal functions'])\n", "('method used for task', 'in addition , to evaluate achievement of the proposed method , the lfpso algorithm is compared the other XXXXX and other methods . the lfpso outperforms the XXXXX and other algorithms and it is closely successful with the XXXXX', ['abc algorithm', 'pso variants'])\n", "('method used for task', 'in addition , to evaluate achievement of the proposed method , the lfpso algorithm is compared the other XXXXX and other methods . the lfpso outperforms the XXXXX and other algorithms and it is closely successful with the XXXXX', ['abc algorithm', 'pso variants'])\n", "('method used for task', 'grnn is proposed to modify XXXXX and XXXXX decomposed pitch residuals', ['vocal tract', 'wavelet packet'])\n", "('NONE', 'XXXXX of grnn reduces XXXXX and overtraining of conventional ann', ['fast convergence', 'computation time'])\n", "('method used for task', 'we propose two novel evolutionary XXXXX approaches ( seann and teann ) for XXXXX ( tsf )', ['neural network', 'time series forecasting'])\n", "('NONE', 'for the weighted aggregation method , give a search method for finding XXXXX located in concave region on the XXXXX', ['pareto-optimal solutions', 'pareto front'])\n", "('method used for task', 'for the weighted aggregation method , give a XXXXX for finding XXXXX located in concave region on the pareto front', ['pareto-optimal solutions', 'search method'])\n", "('method used for task', 'for the weighted aggregation method , give a XXXXX for finding pareto-optimal solutions located in concave region on the XXXXX', ['pareto front', 'search method'])\n", "('method used for task', 'a hybridization of the XXXXX with nelder–mead method to produce an effective and efficient XXXXX', ['global optimization', 'search algorithm'])\n", "('method used for task', 'a set of XXXXX on synthetic signals shows the good performance with respect to fuzzy piecewise linear XXXXX', ['computational experiments', 'regression models'])\n", "('method used for task', 'a combined XXXXX and gradient-based optimization algorithm is applied for automatic selection of proper input variables and the model-dependent variable , and optimizing the XXXXX simultaneously', ['genetic optimization', 'model parameters'])\n", "('method used for task', 'a combined XXXXX and gradient-based optimization algorithm is applied for automatic selection of proper XXXXX and the model-dependent variable , and optimizing the model parameters simultaneously', ['genetic optimization', 'input variables'])\n", "('method used for task', 'a combined genetic optimization and gradient-based optimization algorithm is applied for automatic selection of proper XXXXX and the model-dependent variable , and optimizing the XXXXX simultaneously', ['model parameters', 'input variables'])\n", "('method used for task', 'XXXXX is applied to underwater glider XXXXX', ['differential evolution algorithm', 'path planning'])\n", "('method used for task', 'XXXXX for XXXXX', ['group decision making', 'medical diagnosis'])\n", "('NONE', 'XXXXX of fuzzy uncertain beam equation using XXXXX subject to unit step and impulse loads', ['numerical solution', 'adomian decomposition method'])\n", "('NONE', 'fuzziness appeared in the XXXXX are modeled through convex normalized XXXXX viz . triangular fuzzy numbers', ['initial conditions', 'fuzzy sets'])\n", "('NONE', 'fuzziness appeared in the XXXXX are modeled through convex normalized fuzzy sets viz . XXXXX', ['initial conditions', 'triangular fuzzy numbers'])\n", "('NONE', 'fuzziness appeared in the initial conditions are modeled through convex normalized XXXXX viz . XXXXX', ['fuzzy sets', 'triangular fuzzy numbers'])\n", "('method used for task', 'XXXXX ( adm ) and double parametric form is used with fuzzy based approach to obtain the uncertain bounds of the XXXXX', ['adomian decomposition method', 'dynamic responses'])\n", "('method used for task', 'classification using XXXXX is 99 % or more with near zero XXXXX', ['random forests', 'error measures'])\n", "('NONE', 'the present study analyzes the XXXXX of a repairable industrial system utilizing XXXXX', ['fuzzy reliability', 'uncertain data'])\n", "('method used for task', 'the problem is that when XXXXX is too large , XXXXX can not be used', ['state space', 'model checking'])\n", "('method used for task', 'we suggest using XXXXX to avoid searching the entire XXXXX', ['genetic algorithm', 'state space'])\n", "('NONE', 'the approach involves XXXXX or particle swarm XXXXX', ['simulated annealing', 'optimization algorithms'])\n", "('method used for task', 'we propose a novel hybrid XXXXX ( hvns ) algorithm for XXXXX ( hfs ) scheduling problems', ['variable neighborhood search', 'hybrid flow shop'])\n", "('method used for task', 'we propose a novel hybrid XXXXX ( hvns ) algorithm for hybrid flow shop ( hfs ) XXXXX', ['variable neighborhood search', 'scheduling problems'])\n", "('NONE', 'we propose a novel hybrid variable neighborhood search ( hvns ) algorithm for XXXXX ( hfs ) XXXXX', ['hybrid flow shop', 'scheduling problems'])\n", "('NONE', 'considering the XXXXX , eight XXXXX are developed', ['neighborhood structures', 'problem structure'])\n", "('method used for task', 'address the potential of XXXXX to forecast ground-level ozone in XXXXX', ['learning machine', 'urban area'])\n", "('method used for task', 'XXXXX of parameters for igsa is discussed on XXXXX', ['sensitivity analysis', 'test functions'])\n", "('method used for task', 'the overall approach of dnsa is to update the XXXXX dynamically and repair its XXXXX', ['graph model', 'spanning tree'])\n", "('NONE', 'XXXXX where gray coding is applied , improves the performance of the algorithm in XXXXX', ['parallel processing', 'large-scale problems'])\n", "('method used for task', 'XXXXX are employed to tuning the XXXXX', ['optimization algorithms', 'controller parameters'])\n", "('method used for task', 'a XXXXX is proposed to control a nonlinear XXXXX', ['hybrid method', 'dynamic system'])\n", "('NONE', 'this XXXXX combines XXXXX and kohonen algorithm to obtain faster convergence', ['hybrid algorithm', 'gradient method'])\n", "('method used for task', 'a XXXXX is designed by combining the ga , rbf-nn and XXXXX approaches', ['hybrid model', 'sugeno fuzzy logic'])\n", "('NONE', 'the rbf-nn is used to enhance the pid parameters obtained from ga to design sugeno XXXXX tuned by XXXXX ( ke , τe )', ['fuzzy pid controller', 'excitation parameter'])\n", "('method used for task', 'XXXXX based on XXXXX is designed to solve the proposed model', ['fuzzy simulation', 'tabu search algorithm'])\n", "('NONE', 'this paper proposes a new approach for training XXXXX with a XXXXX determination system', ['support vector machines', 'bone age'])\n", "('method used for task', 'the proposed approach is a combination of XXXXX ( pso ) and XXXXX ( svms )', ['particle swarm optimization', 'support vector machines'])\n", "('NONE', 'for comparison purposes , 6 nonlinear regression , 6 XXXXX , 5 anfis , 3 online XXXXX denfis , 3 offline XXXXX denfis , and 3 offline high-order tsk denfis models were developed', ['neural network', 'first-order tsk'])\n", "('NONE', 'for comparison purposes , 6 nonlinear regression , 6 XXXXX , 5 anfis , 3 online XXXXX denfis , 3 offline XXXXX denfis , and 3 offline high-order tsk denfis models were developed', ['neural network', 'first-order tsk'])\n", "('NONE', 'for comparison purposes , 6 nonlinear regression , 6 neural network , 5 anfis , 3 online XXXXX denfis , 3 offline XXXXX denfis , and 3 offline high-order tsk denfis models were developed', ['first-order tsk', 'first-order denfis'])\n", "('NONE', 'for comparison purposes , 6 nonlinear regression , 6 neural network , 5 anfis , 3 online XXXXX denfis , 3 offline XXXXX denfis , and 3 offline high-order tsk denfis models were developed', ['first-order tsk', 'first-order denfis'])\n", "('NONE', 'for comparison purposes , 6 nonlinear regression , 6 neural network , 5 anfis , 3 online XXXXX denfis , 3 offline XXXXX denfis , and 3 offline high-order tsk denfis models were developed', ['first-order tsk', 'high-order denfis'])\n", "('NONE', 'for comparison purposes , 6 nonlinear regression , 6 neural network , 5 anfis , 3 online XXXXX denfis , 3 offline XXXXX denfis , and 3 offline high-order tsk denfis models were developed', ['first-order denfis', 'first-order tsk'])\n", "('NONE', 'for comparison purposes , 6 nonlinear regression , 6 neural network , 5 anfis , 3 online XXXXX denfis , 3 offline XXXXX denfis , and 3 offline high-order tsk denfis models were developed', ['first-order tsk', 'first-order denfis'])\n", "('NONE', 'for comparison purposes , 6 nonlinear regression , 6 neural network , 5 anfis , 3 online XXXXX denfis , 3 offline XXXXX denfis , and 3 offline high-order tsk denfis models were developed', ['first-order tsk', 'high-order denfis'])\n", "('method used for task', 'XXXXX model could be trained to produce more reliable prediction results in comparison with XXXXX , anfis and nonlinear regression models', ['neural network', 'high-order denfis'])\n", "('NONE', 'a new fractal dimensional XXXXX on XXXXX is proposed', ['model based', 'fuzzy sets theory'])\n", "('NONE', 'we propose a tlbo algorithm for XXXXX of x-bar XXXXX', ['economic design', 'control chart'])\n", "('method used for task', 'we propose a XXXXX for XXXXX of x-bar control chart', ['economic design', 'tlbo algorithm'])\n", "('NONE', 'we propose a XXXXX for economic design of x-bar XXXXX', ['control chart', 'tlbo algorithm'])\n", "('method used for task', 'results of XXXXX are helpful to quality engineers in identifying the significant cost and XXXXX', ['sensitivity analysis', 'process parameters'])\n", "('NONE', 'tlbo is a promising method for the XXXXX of x-bar XXXXX', ['economic design', 'control chart'])\n", "('method used for task', 'a learning-guided XXXXX for constrained XXXXX is proposed', ['multi-objective evolutionary algorithm', 'portfolio optimization problem'])\n", "('method used for task', 'an extended version of weighted aggregated XXXXX assessment ( waspas method ) is proposed for XXXXX', ['soft computing', 'sum product'])\n", "('NONE', 'this study proposes a new method for XXXXX utilizing pso combined with a XXXXX', ['gene selection', 'decision tree'])\n", "('NONE', 'the approach is able to solve XXXXX when XXXXX on the number of clusters is not available', ['data clustering', 'prior knowledge'])\n", "('NONE', 'we have employed XXXXX , XXXXX , k-nearest neighbors , and support vector machine to improve churn prediction', ['decision tree', 'artificial neural networks'])\n", "('NONE', 'we have employed XXXXX , artificial neural networks , XXXXX , and support vector machine to improve churn prediction', ['decision tree', 'k-nearest neighbors'])\n", "('method used for task', 'we have employed XXXXX , artificial neural networks , k-nearest neighbors , and XXXXX to improve churn prediction', ['decision tree', 'support vector machine'])\n", "('method used for task', 'we have employed XXXXX , artificial neural networks , k-nearest neighbors , and support vector machine to improve XXXXX', ['decision tree', 'churn prediction'])\n", "('NONE', 'we have employed decision tree , XXXXX , XXXXX , and support vector machine to improve churn prediction', ['artificial neural networks', 'k-nearest neighbors'])\n", "('method used for task', 'we have employed decision tree , XXXXX , k-nearest neighbors , and XXXXX to improve churn prediction', ['artificial neural networks', 'support vector machine'])\n", "('method used for task', 'we have employed decision tree , XXXXX , k-nearest neighbors , and support vector machine to improve XXXXX', ['artificial neural networks', 'churn prediction'])\n", "('method used for task', 'we have employed decision tree , artificial neural networks , XXXXX , and XXXXX to improve churn prediction', ['k-nearest neighbors', 'support vector machine'])\n", "('method used for task', 'we have employed decision tree , artificial neural networks , XXXXX , and support vector machine to improve XXXXX', ['k-nearest neighbors', 'churn prediction'])\n", "('method used for task', 'we have employed decision tree , artificial neural networks , k-nearest neighbors , and XXXXX to improve XXXXX', ['support vector machine', 'churn prediction'])\n", "('NONE', 'the aim of this paper is to solve the problem of decision makings by introducing soft XXXXX in XXXXX', ['discernibility matrix', 'soft sets'])\n", "('method used for task', 'the notion of soft XXXXX is firstly introduced in XXXXX', ['discernibility matrix', 'soft sets'])\n", "('method used for task', 'an novel algorithm based on the soft XXXXX is proposed to solve the problems of XXXXX', ['discernibility matrix', 'decision making'])\n", "('method used for task', 'the weighted soft XXXXX is introduced in XXXXX and its application to decision making is also investigated', ['discernibility matrix', 'soft sets'])\n", "('method used for task', 'the weighted soft XXXXX is introduced in soft sets and its application to XXXXX is also investigated', ['discernibility matrix', 'decision making'])\n", "('method used for task', 'the weighted soft discernibility matrix is introduced in XXXXX and its application to XXXXX is also investigated', ['soft sets', 'decision making'])\n", "('NONE', 'converting the XXXXX into XXXXX matrices', ['stability problem', 'system parameter'])\n", "('NONE', 'XXXXX within an interval-valued XXXXX', ['multiple criteria decision analysis', 'fuzzy environment'])\n", "('method used for task', 'a XXXXX is implemented with computational XXXXX', ['comparative study', 'experimental analysis'])\n", "('method used for task', 'presenting self-tuning XXXXX based on XXXXX emotional learning to control dvr', ['pi controller', 'human brain'])\n", "('method used for task', 'XXXXX and final answer are great better in tlbo algorithm than XXXXX', ['convergence speed', 'pso algorithm'])\n", "('method used for task', 'XXXXX and final answer are great better in XXXXX than pso algorithm', ['convergence speed', 'tlbo algorithm'])\n", "('NONE', 'convergence speed and final answer are great better in XXXXX than XXXXX', ['pso algorithm', 'tlbo algorithm'])\n", "('NONE', 'the proposed XXXXX enhances the achieved strip thickness under high XXXXX', ['interval type-2 fuzzy logic system', 'uncertainty level'])\n", "('NONE', 'we take covering XXXXX as models of e-spread XXXXX', ['approximation spaces', 'information systems'])\n", "('method used for task', 'we solved the XXXXX for different cases and different XXXXX', ['optimal power flow', 'test systems'])\n", "('NONE', 'bhbo is conceptually very simple , further unlike other XXXXX bhbo parameter-less XXXXX', ['optimization techniques', 'optimization technique'])\n", "('method used for task', 'XXXXX for XXXXX', ['wireless sensor networks', 'soft computing techniques'])\n", "('NONE', 'we utilize XXXXX ( XXXXX and classification ) on ct images', ['image analysis', 'feature extraction'])\n", "('method used for task', 'we utilize XXXXX ( feature extraction and classification ) on XXXXX', ['image analysis', 'ct images'])\n", "('method used for task', 'we utilize image analysis ( XXXXX and classification ) on XXXXX', ['feature extraction', 'ct images'])\n", "('method used for task', 'an evolutionary XXXXX was implemented to develop a XXXXX on a comprehensive field measurements database', ['model based', 'data mining technique'])\n", "('method used for task', 'neutrosophic XXXXX is defined to describe the XXXXX on images', ['similarity function', 'uncertain information'])\n", "('method used for task', 'a novel XXXXX is defined using neutrosophic XXXXX and the new defined clustering algorithm classifies the pixels on the image into different groups', ['objective function', 'similarity function'])\n", "('method used for task', 'the proposed decision framework is able to derive XXXXX for building cluster which could significantly reduce XXXXX', ['pareto solutions', 'energy cost'])\n", "('method used for task', 'the proposed XXXXX is able to derive XXXXX for building cluster which could significantly reduce energy cost', ['pareto solutions', 'decision framework'])\n", "('method used for task', 'the proposed XXXXX is able to derive pareto solutions for building cluster which could significantly reduce XXXXX', ['energy cost', 'decision framework'])\n", "('NONE', 'compare to a XXXXX based decision framework , the proposed decision framework could significantly reduce XXXXX', ['memetic algorithm', 'computational cost'])\n", "('NONE', 'compare to a XXXXX based XXXXX , the proposed XXXXX could significantly reduce computational cost', ['memetic algorithm', 'decision framework'])\n", "('NONE', 'compare to a XXXXX based XXXXX , the proposed XXXXX could significantly reduce computational cost', ['memetic algorithm', 'decision framework'])\n", "('NONE', 'compare to a memetic algorithm based XXXXX , the proposed XXXXX could significantly reduce XXXXX', ['computational cost', 'decision framework'])\n", "('NONE', 'compare to a memetic algorithm based XXXXX , the proposed XXXXX could significantly reduce XXXXX', ['computational cost', 'decision framework'])\n", "('method used for task', 'we introduce a novel XXXXX to finding the XXXXX of a nonlinear function', ['iterative method', 'fixed point'])\n", "('method used for task', 'we combine ideas proposed in artificial bee XXXXX and XXXXX', ['bisection method', 'colony algorithm'])\n", "('NONE', 'it considers another uncertainty in the XXXXX of XXXXX', ['membership function', 'fuzzy set'])\n", "('NONE', 'the XXXXX is one of the leading XXXXX', ['multiobjective evolutionary algorithms', 'moea/d algorithm'])\n", "('NONE', 'XXXXX is a branch of XXXXX in which the agent must learn through interaction with the environment', ['reinforcement learning', 'machine learning'])\n", "('method used for task', 'XXXXX and XXXXX are conducted as a unifying model selection', ['feature selection', 'parameter tuning'])\n", "('method used for task', 'XXXXX and parameter tuning are conducted as a unifying XXXXX', ['feature selection', 'model selection'])\n", "('NONE', 'feature selection and XXXXX are conducted as a unifying XXXXX', ['parameter tuning', 'model selection'])\n", "('method used for task', 'we present a study about the performance of two classical weight training methods of XXXXX ( rbfn ) , least mean square ( lms ) and XXXXX ( svd ) , applied to classification problems , when the data-sets are imbalanced', ['radial basis function networks', 'singular value decomposition'])\n", "('method used for task', 'we present a study about the performance of two classical weight training methods of XXXXX ( rbfn ) , least mean square ( lms ) and singular value decomposition ( svd ) , applied to XXXXX , when the data-sets are imbalanced', ['radial basis function networks', 'classification problems'])\n", "('method used for task', 'we present a study about the performance of two classical weight training methods of radial basis function networks ( rbfn ) , least mean square ( lms ) and XXXXX ( svd ) , applied to XXXXX , when the data-sets are imbalanced', ['singular value decomposition', 'classification problems'])\n", "('NONE', 'XXXXX permitted the analysis of the behavior of our algorithm for different categories of XXXXX', ['experimental evaluation', 'mobile applications'])\n", "('method used for task', 'we present a review paper on the importance of data granulation in the wide contexts of XXXXX and XXXXX', ['soft computing', 'pattern recognition'])\n", "('NONE', 'we present a XXXXX on the importance of data granulation in the wide contexts of XXXXX and pattern recognition', ['soft computing', 'review paper'])\n", "('NONE', 'we present a XXXXX on the importance of data granulation in the wide contexts of soft computing and XXXXX', ['pattern recognition', 'review paper'])\n", "('NONE', 'we extend the topsis approach to rank all the alternatives from the perspective of the magnitude of XXXXX under interval-valued intuitionistic XXXXX', ['fuzzy environment', 'decision information'])\n", "('NONE', 'analytic XXXXX ( ahp ) was proposed to achieve weights of criteria in a XXXXX', ['hierarchy process', 'group decision-making problem'])\n", "('NONE', 'the XXXXX could be provided by the user or could be the result of performing a conventional XXXXX over the input data', ['clustering method', 'label information'])\n", "('NONE', 'we develop XXXXX ( ann , fis , and anfis ) for estimating the number of XXXXX based on the occurrence of economic improvement projects', ['adverse events', 'soft computing techniques'])\n", "('NONE', 'we develop soft computing techniques ( ann , fis , and anfis ) for estimating the number of XXXXX based on the occurrence of economic XXXXX', ['adverse events', 'improvement projects'])\n", "('NONE', 'we develop XXXXX ( ann , fis , and anfis ) for estimating the number of adverse events based on the occurrence of economic XXXXX', ['soft computing techniques', 'improvement projects'])\n", "('NONE', 'when the XXXXX was calculated based on the mape for each of the models , ann had better XXXXX than fis and anfis models , as demonstrated by experimental results', ['model accuracy', 'predictive accuracy'])\n", "('NONE', 'in this paper we introduce an adapted elitist XXXXX for automatic knot adjustment of XXXXX', ['clonal selection algorithm', 'b-spline curves'])\n", "('method used for task', 'adaptive XXXXX ( agfs ) for optimizing rules and XXXXX', ['genetic fuzzy system', 'membership functions'])\n", "('NONE', 'we address the problem of XXXXX in multidimensional XXXXX', ['missing data', 'time series'])\n", "('method used for task', 'de novo XXXXX supplies novel molecules for XXXXX', ['drug design', 'drug development'])\n", "('method used for task', 'XXXXX are used for XXXXX', ['evolutionary algorithms', 'multi-objective optimization'])\n", "('NONE', 'we create XXXXX ( fis ) as a means of computerizing XXXXX ( dd ) tables', ['fuzzy inference systems', 'differential diagnosis'])\n", "('method used for task', 'XXXXX based on a simulator are performed to validate the effectiveness of the proposed XXXXX', ['numerical experiments', 'decision rules'])\n", "('NONE', 'since the key space of vigenere cryptosystem is very large , therefore , brute force does not help in the analysis of the cryptosystem in XXXXX . XXXXX are also not helpful when the key size is not small', ['real time', 'statistical techniques'])\n", "('NONE', 'the aim of this work was to examine the feasibility of XXXXX as a cryptanalysis tool of classical cryptosystems especially vigenere cipher . we applied XXXXX , particle swarm optimization and cuckoo search techniques for the analysis of vigenere cipher . after doing lots of experiments we observed that ga and pso techniques can recover the entire key of vigenere cipher correctly for keys of small lengths . however , cuckoo search can recover more than 90 % of the key characters for keys of size up to 25 characters because ga and pso get trapped in local minima while cuckoo search finds the global optimum solution following characteristics of lévy flights', ['cuckoo search algorithm', 'genetic algorithm'])\n", "('NONE', 'the aim of this work was to examine the feasibility of XXXXX as a cryptanalysis tool of classical cryptosystems especially vigenere cipher . we applied genetic algorithm , XXXXX and cuckoo search techniques for the analysis of vigenere cipher . after doing lots of experiments we observed that ga and pso techniques can recover the entire key of vigenere cipher correctly for keys of small lengths . however , cuckoo search can recover more than 90 % of the key characters for keys of size up to 25 characters because ga and pso get trapped in local minima while cuckoo search finds the global optimum solution following characteristics of lévy flights', ['cuckoo search algorithm', 'particle swarm optimization'])\n", "('NONE', 'the aim of this work was to examine the feasibility of XXXXX as a cryptanalysis tool of classical cryptosystems especially vigenere cipher . we applied genetic algorithm , particle swarm optimization and XXXXX techniques for the analysis of vigenere cipher . after doing lots of experiments we observed that ga and pso techniques can recover the entire key of vigenere cipher correctly for keys of small lengths . however , XXXXX can recover more than 90 % of the key characters for keys of size up to 25 characters because ga and pso get trapped in local minima while XXXXX finds the global optimum solution following characteristics of lévy flights', ['cuckoo search algorithm', 'cuckoo search'])\n", "('NONE', 'the aim of this work was to examine the feasibility of XXXXX as a cryptanalysis tool of classical cryptosystems especially vigenere cipher . we applied genetic algorithm , particle swarm optimization and cuckoo search techniques for the analysis of vigenere cipher . after doing lots of experiments we observed that ga and pso techniques can recover the entire key of vigenere cipher correctly for keys of small lengths . however , cuckoo search can recover more than 90 % of the key characters for keys of size up to 25 characters because ga and pso get trapped in XXXXX while cuckoo search finds the global optimum solution following characteristics of lévy flights', ['cuckoo search algorithm', 'local minima'])\n", "('NONE', 'the aim of this work was to examine the feasibility of XXXXX as a cryptanalysis tool of classical cryptosystems especially vigenere cipher . we applied genetic algorithm , particle swarm optimization and XXXXX techniques for the analysis of vigenere cipher . after doing lots of experiments we observed that ga and pso techniques can recover the entire key of vigenere cipher correctly for keys of small lengths . however , XXXXX can recover more than 90 % of the key characters for keys of size up to 25 characters because ga and pso get trapped in local minima while XXXXX finds the global optimum solution following characteristics of lévy flights', ['cuckoo search algorithm', 'cuckoo search'])\n", "('NONE', 'the aim of this work was to examine the feasibility of cuckoo search algorithm as a cryptanalysis tool of classical cryptosystems especially vigenere cipher . we applied XXXXX , XXXXX and cuckoo search techniques for the analysis of vigenere cipher . after doing lots of experiments we observed that ga and pso techniques can recover the entire key of vigenere cipher correctly for keys of small lengths . however , cuckoo search can recover more than 90 % of the key characters for keys of size up to 25 characters because ga and pso get trapped in local minima while cuckoo search finds the global optimum solution following characteristics of lévy flights', ['genetic algorithm', 'particle swarm optimization'])\n", "('NONE', 'the aim of this work was to examine the feasibility of XXXXX algorithm as a cryptanalysis tool of classical cryptosystems especially vigenere cipher . we applied XXXXX , particle swarm optimization and XXXXX techniques for the analysis of vigenere cipher . after doing lots of experiments we observed that ga and pso techniques can recover the entire key of vigenere cipher correctly for keys of small lengths . however , XXXXX can recover more than 90 % of the key characters for keys of size up to 25 characters because ga and pso get trapped in local minima while XXXXX finds the global optimum solution following characteristics of lévy flights', ['genetic algorithm', 'cuckoo search'])\n", "('NONE', 'the aim of this work was to examine the feasibility of cuckoo search algorithm as a cryptanalysis tool of classical cryptosystems especially vigenere cipher . we applied XXXXX , particle swarm optimization and cuckoo search techniques for the analysis of vigenere cipher . after doing lots of experiments we observed that ga and pso techniques can recover the entire key of vigenere cipher correctly for keys of small lengths . however , cuckoo search can recover more than 90 % of the key characters for keys of size up to 25 characters because ga and pso get trapped in XXXXX while cuckoo search finds the global optimum solution following characteristics of lévy flights', ['genetic algorithm', 'local minima'])\n", "('NONE', 'the aim of this work was to examine the feasibility of XXXXX algorithm as a cryptanalysis tool of classical cryptosystems especially vigenere cipher . we applied XXXXX , particle swarm optimization and XXXXX techniques for the analysis of vigenere cipher . after doing lots of experiments we observed that ga and pso techniques can recover the entire key of vigenere cipher correctly for keys of small lengths . however , XXXXX can recover more than 90 % of the key characters for keys of size up to 25 characters because ga and pso get trapped in local minima while XXXXX finds the global optimum solution following characteristics of lévy flights', ['genetic algorithm', 'cuckoo search'])\n", "('NONE', 'the aim of this work was to examine the feasibility of XXXXX algorithm as a cryptanalysis tool of classical cryptosystems especially vigenere cipher . we applied genetic algorithm , XXXXX and XXXXX techniques for the analysis of vigenere cipher . after doing lots of experiments we observed that ga and pso techniques can recover the entire key of vigenere cipher correctly for keys of small lengths . however , XXXXX can recover more than 90 % of the key characters for keys of size up to 25 characters because ga and pso get trapped in local minima while XXXXX finds the global optimum solution following characteristics of lévy flights', ['particle swarm optimization', 'cuckoo search'])\n", "('NONE', 'the aim of this work was to examine the feasibility of cuckoo search algorithm as a cryptanalysis tool of classical cryptosystems especially vigenere cipher . we applied genetic algorithm , XXXXX and cuckoo search techniques for the analysis of vigenere cipher . after doing lots of experiments we observed that ga and pso techniques can recover the entire key of vigenere cipher correctly for keys of small lengths . however , cuckoo search can recover more than 90 % of the key characters for keys of size up to 25 characters because ga and pso get trapped in XXXXX while cuckoo search finds the global optimum solution following characteristics of lévy flights', ['particle swarm optimization', 'local minima'])\n", "('NONE', 'the aim of this work was to examine the feasibility of XXXXX algorithm as a cryptanalysis tool of classical cryptosystems especially vigenere cipher . we applied genetic algorithm , XXXXX and XXXXX techniques for the analysis of vigenere cipher . after doing lots of experiments we observed that ga and pso techniques can recover the entire key of vigenere cipher correctly for keys of small lengths . however , XXXXX can recover more than 90 % of the key characters for keys of size up to 25 characters because ga and pso get trapped in local minima while XXXXX finds the global optimum solution following characteristics of lévy flights', ['particle swarm optimization', 'cuckoo search'])\n", "('NONE', 'the aim of this work was to examine the feasibility of XXXXX algorithm as a cryptanalysis tool of classical cryptosystems especially vigenere cipher . we applied genetic algorithm , particle swarm optimization and XXXXX techniques for the analysis of vigenere cipher . after doing lots of experiments we observed that ga and pso techniques can recover the entire key of vigenere cipher correctly for keys of small lengths . however , XXXXX can recover more than 90 % of the key characters for keys of size up to 25 characters because ga and pso get trapped in XXXXX while XXXXX finds the global optimum solution following characteristics of lévy flights', ['cuckoo search', 'local minima'])\n", "('NONE', 'the aim of this work was to examine the feasibility of XXXXX algorithm as a cryptanalysis tool of classical cryptosystems especially vigenere cipher . we applied genetic algorithm , particle swarm optimization and XXXXX techniques for the analysis of vigenere cipher . after doing lots of experiments we observed that ga and pso techniques can recover the entire key of vigenere cipher correctly for keys of small lengths . however , XXXXX can recover more than 90 % of the key characters for keys of size up to 25 characters because ga and pso get trapped in XXXXX while XXXXX finds the global optimum solution following characteristics of lévy flights', ['local minima', 'cuckoo search'])\n", "('NONE', \"idsfq is presented to provide adaptive control scenario for dynamic XXXXX inside cloud 's XXXXX\", ['resource provisioning', 'spot market'])\n", "('NONE', 'a new XXXXX of the protein folding problem in 2d-hp model is proposed for the use of XXXXX', ['state space representation', 'reinforcement learning methods'])\n", "('NONE', 'the proposed XXXXX also provides an actual learning for an agent . thus , at the end of a XXXXX an agent could find the optimum fold of any sequence of a certain length , which is not the case in the existing reinforcement learning methods', ['state space representation', 'learning process'])\n", "('NONE', 'the proposed XXXXX also provides an actual learning for an agent . thus , at the end of a learning process an agent could find the optimum fold of any sequence of a certain length , which is not the case in the existing XXXXX', ['state space representation', 'reinforcement learning methods'])\n", "('NONE', 'the proposed state space representation also provides an actual learning for an agent . thus , at the end of a XXXXX an agent could find the optimum fold of any sequence of a certain length , which is not the case in the existing XXXXX', ['learning process', 'reinforcement learning methods'])\n", "('method used for task', 'by using the ant-q algorithm ( an ant based reinforcement learning method ) , optimum fold of a XXXXX is found rapidly when compared to the standard XXXXX', ['protein sequence', 'q-learning algorithm'])\n", "('method used for task', 'a simpler approach to XXXXX based on XXXXX is presented with covering based rough sets', ['attribute reduction', 'discernibility matrix'])\n", "('method used for task', 'a simpler approach to XXXXX based on discernibility matrix is presented with covering based XXXXX', ['attribute reduction', 'rough sets'])\n", "('method used for task', 'a XXXXX to XXXXX based on discernibility matrix is presented with covering based rough sets', ['attribute reduction', 'simpler approach'])\n", "('method used for task', 'a simpler approach to attribute reduction based on XXXXX is presented with covering based XXXXX', ['discernibility matrix', 'rough sets'])\n", "('method used for task', 'a XXXXX to attribute reduction based on XXXXX is presented with covering based rough sets', ['discernibility matrix', 'simpler approach'])\n", "('method used for task', 'a XXXXX to attribute reduction based on discernibility matrix is presented with covering based XXXXX', ['rough sets', 'simpler approach'])\n", "('NONE', 'some important properties of XXXXX with covering based XXXXX are improved', ['attribute reduction', 'rough sets'])\n", "('NONE', 'a new algorithm to XXXXX in XXXXX is presented in a different strategy of identifying objects', ['attribute reduction', 'decision tables'])\n", "('NONE', 'design of three unsupervised XXXXX that satisfying exactly XXXXX', ['initial conditions', 'ann models'])\n", "('method used for task', 'a XXXXX based on modified servqual model and XXXXX is used', ['hybrid method', 'fuzzy set theory'])\n", "('method used for task', 'improvement in performance by asset utilization , XXXXX and XXXXX in terms of production loss', ['energy efficient', 'cost reduction'])\n", "('method used for task', 'a case based reasoning system with an original XXXXX for biomedical XXXXX is proposed', ['hidden markov model', 'text classification'])\n", "('NONE', 'the model of XXXXX with XXXXX and the seasonal mechanism is proposed', ['support vector regression', 'adaptive genetic algorithm'])\n", "('method used for task', 'XXXXX ( rl ) is applied to construct the XXXXX', ['reinforcement learning', 'probabilistic model'])\n", "('method used for task', 'this paper demonstrates that the recently discovered XXXXX is a better technique for the optimization of large XXXXX', ['firefly algorithm', 'wind farms'])\n", "('NONE', 'it compares the XXXXX with the XXXXX or by the use of spread sheet methods such as finite difference method', ['firefly algorithm', 'genetic algorithms'])\n", "('NONE', 'it compares the XXXXX with the genetic algorithms or by the use of spread sheet methods such as XXXXX', ['firefly algorithm', 'finite difference method'])\n", "('NONE', 'it compares the firefly algorithm with the XXXXX or by the use of spread sheet methods such as XXXXX', ['genetic algorithms', 'finite difference method'])\n", "('NONE', 'this paper demonstrates that the XXXXX outperforms the XXXXX and finite difference method by a wide margin', ['firefly algorithm', 'genetic algorithms'])\n", "('NONE', 'this paper demonstrates that the XXXXX outperforms the genetic algorithms and XXXXX by a wide margin', ['firefly algorithm', 'finite difference method'])\n", "('method used for task', 'this paper demonstrates that the firefly algorithm outperforms the XXXXX and XXXXX by a wide margin', ['genetic algorithms', 'finite difference method'])\n", "('NONE', 'the crack growth rate has been calculated by using an XXXXX from experimental crack length ( a ) and number of cycles ( n ) data which has been subsequently used as training data base for gp XXXXX along with load ratio ( r ) , maximum stress intensity factor ( kmax ) and stress intensity factor rage ( δk )', ['exponential model', 'model formulation'])\n", "('NONE', 'the crack growth rate has been calculated by using an XXXXX from experimental crack length ( a ) and number of cycles ( n ) data which has been subsequently used as training data base for gp model formulation along with load ratio ( r ) , maximum XXXXX ( kmax ) and XXXXX rage ( δk )', ['exponential model', 'stress intensity factor'])\n", "('NONE', 'the crack growth rate has been calculated by using an XXXXX from experimental crack length ( a ) and number of cycles ( n ) data which has been subsequently used as training data base for gp model formulation along with load ratio ( r ) , maximum XXXXX ( kmax ) and XXXXX rage ( δk )', ['exponential model', 'stress intensity factor'])\n", "('NONE', 'the XXXXX has been calculated by using an XXXXX from experimental crack length ( a ) and number of cycles ( n ) data which has been subsequently used as training data base for gp model formulation along with load ratio ( r ) , maximum stress intensity factor ( kmax ) and stress intensity factor rage ( δk )', ['exponential model', 'crack growth rate'])\n", "('NONE', 'the crack growth rate has been calculated by using an exponential model from experimental crack length ( a ) and number of cycles ( n ) data which has been subsequently used as training data base for gp XXXXX along with load ratio ( r ) , maximum XXXXX ( kmax ) and XXXXX rage ( δk )', ['model formulation', 'stress intensity factor'])\n", "('NONE', 'the crack growth rate has been calculated by using an exponential model from experimental crack length ( a ) and number of cycles ( n ) data which has been subsequently used as training data base for gp XXXXX along with load ratio ( r ) , maximum XXXXX ( kmax ) and XXXXX rage ( δk )', ['model formulation', 'stress intensity factor'])\n", "('NONE', 'the XXXXX has been calculated by using an exponential model from experimental crack length ( a ) and number of cycles ( n ) data which has been subsequently used as training data base for gp XXXXX along with load ratio ( r ) , maximum stress intensity factor ( kmax ) and stress intensity factor rage ( δk )', ['model formulation', 'crack growth rate'])\n", "('NONE', 'the XXXXX has been calculated by using an exponential model from experimental crack length ( a ) and number of cycles ( n ) data which has been subsequently used as training data base for gp model formulation along with load ratio ( r ) , maximum XXXXX ( kmax ) and XXXXX rage ( δk )', ['stress intensity factor', 'crack growth rate'])\n", "('NONE', 'the XXXXX has been calculated by using an exponential model from experimental crack length ( a ) and number of cycles ( n ) data which has been subsequently used as training data base for gp model formulation along with load ratio ( r ) , maximum XXXXX ( kmax ) and XXXXX rage ( δk )', ['stress intensity factor', 'crack growth rate'])\n", "('NONE', 'the validity of the proposed gp model has been confirmed by comparing the XXXXX by XXXXX and also with previously proposed ann model', ['experimental data', 'model prediction'])\n", "('NONE', 'the validity of the proposed gp model has been confirmed by comparing the model prediction by XXXXX and also with previously proposed XXXXX', ['experimental data', 'ann model'])\n", "('NONE', 'the validity of the proposed gp model has been confirmed by comparing the XXXXX by experimental data and also with previously proposed XXXXX', ['model prediction', 'ann model'])\n", "('NONE', 'nowadays , software designers attempt to employ XXXXX in software design phase , but XXXXX may be not used in the implementation phase . therefore , one of the challenging issues is XXXXX of source code and design , i.e. , XXXXX', ['design patterns', 'conformance checking'])\n", "('NONE', 'nowadays , software designers attempt to employ XXXXX in software design phase , but XXXXX may be not used in the implementation phase . therefore , one of the challenging issues is conformance checking of XXXXX and design , i.e. , XXXXX', ['design patterns', 'source code'])\n", "('NONE', 'nowadays , software designers attempt to employ XXXXX in software design phase , but XXXXX may be not used in the implementation phase . therefore , one of the challenging issues is XXXXX of source code and design , i.e. , XXXXX', ['design patterns', 'conformance checking'])\n", "('NONE', 'nowadays , software designers attempt to employ XXXXX in software design phase , but XXXXX may be not used in the implementation phase . therefore , one of the challenging issues is conformance checking of XXXXX and design , i.e. , XXXXX', ['design patterns', 'source code'])\n", "('NONE', 'nowadays , software designers attempt to employ design patterns in software design phase , but design patterns may be not used in the implementation phase . therefore , one of the challenging issues is XXXXX of XXXXX and design , i.e. , design patterns', ['conformance checking', 'source code'])\n", "('NONE', 'nowadays , software designers attempt to employ XXXXX in software design phase , but XXXXX may be not used in the implementation phase . therefore , one of the challenging issues is XXXXX of source code and design , i.e. , XXXXX', ['conformance checking', 'design patterns'])\n", "('NONE', 'nowadays , software designers attempt to employ XXXXX in software design phase , but XXXXX may be not used in the implementation phase . therefore , one of the challenging issues is conformance checking of XXXXX and design , i.e. , XXXXX', ['source code', 'design patterns'])\n", "('NONE', 'in addition , after developing a system , usually its documents are not maintained , so , identifying XXXXX from XXXXX can help to achieve the design of an existing system as a reverse engineering task', ['design pattern', 'source code'])\n", "('NONE', 'the variant implementations ( i.e. , different XXXXX ) of a design pattern make hard to detect the design pattern instances from the XXXXX . to address this issue , in this paper , we propose a new method which aims to map the design pattern detection problem into a learning problem', ['source code', 'source codes'])\n", "('method used for task', 'the variant implementations ( i.e. , different source codes ) of a design pattern make hard to detect the design pattern instances from the XXXXX . to address this issue , in this paper , we propose a new method which aims to map the design pattern XXXXX into a learning problem', ['source code', 'detection problem'])\n", "('method used for task', 'the variant implementations ( i.e. , different XXXXX ) of a design pattern make hard to detect the design pattern instances from the source code . to address this issue , in this paper , we propose a new method which aims to map the design pattern XXXXX into a learning problem', ['source codes', 'detection problem'])\n", "('method used for task', 'central XXXXX ( cfo ) is deterministic XXXXX', ['metaheuristic algorithm', 'force optimization'])\n", "('NONE', 'XXXXX are applied for predicting the development effort of XXXXX', ['neural networks', 'software projects'])\n", "('NONE', 'XXXXX of XXXXX is compared with that of a statistical regression', ['prediction accuracy', 'neural networks'])\n", "('NONE', 'a XXXXX had a XXXXX with the highest confidence level', ['radial basis function neural network', 'prediction accuracy'])\n", "('NONE', 'a XXXXX had a prediction accuracy with the highest XXXXX', ['radial basis function neural network', 'confidence level'])\n", "('NONE', 'a radial basis function neural network had a XXXXX with the highest XXXXX', ['prediction accuracy', 'confidence level'])\n", "('method used for task', 'the XXXXX was based on a genetic algorithm-based XXXXX', ['neural fuzzy system', 'software sensor'])\n", "('method used for task', 'fuzzy XXXXX and XXXXX was used to identify the model', ['subtractive clustering', 'genetic algorithm'])\n", "('method used for task', 'we introduce the concept of XXXXX in binary spaces.it is proven that utilizing random numbers and their opposite is beneficial in evolutionary algorithms.opposite numbers are applied to accelerate the XXXXX of binary gravitational search algorithm ( bgsa ) .the results show that obgsa possesses superior performance in accuracy as compared to the bgsa', ['opposition-based learning', 'convergence rate'])\n", "('method used for task', 'we introduce the concept of XXXXX in binary spaces.it is proven that utilizing random numbers and their opposite is beneficial in evolutionary algorithms.opposite numbers are applied to accelerate the convergence rate of binary XXXXX ( bgsa ) .the results show that obgsa possesses superior performance in accuracy as compared to the bgsa', ['opposition-based learning', 'gravitational search algorithm'])\n", "('NONE', 'we introduce the concept of opposition-based learning in binary spaces.it is proven that utilizing random numbers and their opposite is beneficial in evolutionary algorithms.opposite numbers are applied to accelerate the XXXXX of binary XXXXX ( bgsa ) .the results show that obgsa possesses superior performance in accuracy as compared to the bgsa', ['convergence rate', 'gravitational search algorithm'])\n", "('method used for task', 'we reviewed the use of XXXXX on the mdvrp ( multi depot XXXXX )', ['genetic algorithms', 'vehicle routing problem'])\n", "('method used for task', 'we compared the XXXXX to other XXXXX on mdvrp based on the results on standard benchmarks', ['genetic algorithms', 'metaheuristic algorithms'])\n", "('method used for task', 'XXXXX ( pso ) is used to optimize the XXXXX ( svm )', ['particle swarm optimization', 'support vector machine'])\n", "('method used for task', 'the gm ( 1 , n ) model and XXXXX are used for XXXXX , which is capable of simplifying the system prediction model', ['fuzzy c-means clustering', 'variable selection'])\n", "('method used for task', 'proposing acor–pso as a two-stage meta-heuristic XXXXX to efficiently find the XXXXX of double-arch concrete dams', ['optimization model', 'optimal shape'])\n", "('NONE', 'reducing the XXXXX of optimization procedure of double-arch dams subject to earthquake loads using XXXXX', ['computational time', 'weighted least squares support vector machine'])\n", "('method used for task', 'this paper proposes a new method for speaker XXXXX based on formants , wavelet entropy and XXXXX denoted as fwenn', ['feature extraction', 'neural networks'])\n", "('method used for task', 'mapping the XXXXX according to XXXXX', ['image histogram', 'rayleigh distribution'])\n", "('NONE', 'the proposed flow-based XXXXX represents a complete departure from the traditional XXXXX rather than modifies simply the traditional XXXXX', ['similarity measure', 'distance measure'])\n", "('NONE', 'the proposed flow-based XXXXX represents a complete departure from the traditional XXXXX rather than modifies simply the traditional XXXXX', ['similarity measure', 'distance measure'])\n", "('NONE', 'XXXXX performs the following : summarize XXXXX for sfp models', ['systematic literature review', 'ml techniques'])\n", "('NONE', 'assess XXXXX and capability of XXXXX for constructing sfp models', ['performance accuracy', 'ml techniques'])\n", "('NONE', 'provide comparison of XXXXX of different XXXXX', ['performance accuracy', 'ml techniques'])\n", "('NONE', 'we review applications of the XXXXX in the statistical XXXXX', ['time series analysis', 'soft computing techniques'])\n", "('method used for task', 'the employed XXXXX and XXXXX provide useful information for forecasting', ['data mining', 'classification methods'])\n", "('NONE', 'to realize the XXXXX of implementing XXXXX planning system', ['risk factor', 'enterprise resource'])\n", "('NONE', 'a new XXXXX based hybrid XXXXX is proposed to further improve the performance of the algorithm', ['neighborhood structure', 'mutation strategy'])\n", "('method used for task', 'XXXXX is performed under wide changes in XXXXX and disturbance', ['robustness analysis', 'system parameters'])\n", "('NONE', 'based on the introduced XXXXX , two types of XXXXX are proposed whose consequent parts of the rules are linear and quadratic functions', ['membership function', 'ts models'])\n", "('method used for task', 'an XXXXX is suggested to identify the proposed XXXXX', ['incremental learning algorithm', 'ts models'])\n", "('NONE', 'obtained results demonstrate XXXXX and low redundancy of the suggested XXXXX', ['high accuracy', 'ts fuzzy models'])\n", "('method used for task', 'nonconvex emission constrained XXXXX ( neced ) is a complex XXXXX', ['economic dispatch', 'optimization problem'])\n", "('NONE', 'the framework can be used in XXXXX , like XXXXX', ['engineering applications', 'video surveillance'])\n", "('method used for task', 'combination of XXXXX and entropy method is applied for XXXXX weighting', ['fuzzy ahp', 'risk factor'])\n", "('method used for task', 'combination of XXXXX and XXXXX is applied for risk factor weighting', ['fuzzy ahp', 'entropy method'])\n", "('method used for task', 'combination of fuzzy ahp and XXXXX is applied for XXXXX weighting', ['risk factor', 'entropy method'])\n", "('method used for task', 'fuzzy XXXXX is used to determine the risk priorities of XXXXX', ['vikor method', 'failure modes'])\n", "('method used for task', 'this paper also ventures on a new problem on XXXXX compared to previous smaller scale data used in XXXXX of efg in ref . [ 2 ] and text extraction task in refs . [ 1,6 ]', ['text classification', 'case study'])\n", "('NONE', 'we use an improved multiobjective estimation of distribution algorithm ( irm-meda ) to solve the environmental XXXXX of hydrothermal XXXXX', ['economic dispatch', 'power systems'])\n", "('method used for task', 'we use an improved multiobjective estimation of XXXXX ( irm-meda ) to solve the environmental XXXXX of hydrothermal power systems', ['economic dispatch', 'distribution algorithm'])\n", "('method used for task', 'we use an improved multiobjective estimation of XXXXX ( irm-meda ) to solve the environmental economic dispatch of hydrothermal XXXXX', ['power systems', 'distribution algorithm'])\n", "('method used for task', 'the proposed algorithm obtained new optimal solutions and XXXXX for XXXXX', ['upper bounds', 'benchmark problems'])\n", "('NONE', 'a new method of XXXXX of the generalized XXXXX', ['similarity measure', 'trapezoidal fuzzy numbers'])\n", "('method used for task', 'combining ga and tmvdr achieve XXXXX and XXXXX', ['high accuracy', 'fast convergence'])\n", "('NONE', 'to our knowledge , it was the first study in identifying the XXXXX with XXXXX', ['computer vision', 'butterfly species'])\n", "('method used for task', 'the method is based on XXXXX and XXXXX', ['local binary patterns', 'artificial neural network'])\n", "('method used for task', \"a novel XXXXX based on improved culture algorithm for a vehicle 's active suspension was proposed and supported by XXXXX\", ['control strategy', 'numerical simulation'])\n", "('method used for task', \"a novel XXXXX based on improved XXXXX for a vehicle 's active suspension was proposed and supported by numerical simulation\", ['control strategy', 'culture algorithm'])\n", "('method used for task', \"a novel control strategy based on improved XXXXX for a vehicle 's active suspension was proposed and supported by XXXXX\", ['numerical simulation', 'culture algorithm'])\n", "('NONE', 'by simulation , the perfect control effect is obtained by using fuzzy XXXXX with the optimal XXXXX optimized by improved culture algorithm', ['pid control', 'fuzzy rules'])\n", "('NONE', 'by simulation , the perfect control effect is obtained by using fuzzy XXXXX with the optimal fuzzy rules optimized by improved XXXXX', ['pid control', 'culture algorithm'])\n", "('NONE', 'by simulation , the perfect control effect is obtained by using fuzzy pid control with the optimal XXXXX optimized by improved XXXXX', ['fuzzy rules', 'culture algorithm'])\n", "('method used for task', 'proposing a novel framework for evaluating teaching performance based on the combination of XXXXX and fuzzy comprehensive XXXXX', ['fuzzy ahp', 'evaluation method'])\n", "('method used for task', 'the output gain of the adaptive XXXXX ( afc ) , it is considerate as a XXXXX', ['fuzzy controller', 'fuzzy variable'])\n", "('NONE', 'the comparative at XXXXX study , shows that the adaptive XXXXX can be enhance the performances the dc voltage control at various operating', ['pi controller', 'fuzzy controller'])\n", "('method used for task', 'XXXXX is developed by combining XXXXX ( ga ) , radial basis function neural network ( rbf-nn ) and sugeno fuzzy logic approaches', ['integrated model', 'genetic algorithm'])\n", "('method used for task', 'XXXXX is developed by combining genetic algorithm ( ga ) , XXXXX ( rbf-nn ) and sugeno fuzzy logic approaches', ['integrated model', 'radial basis function neural network'])\n", "('method used for task', 'XXXXX is developed by combining genetic algorithm ( ga ) , radial basis function neural network ( rbf-nn ) and XXXXX approaches', ['integrated model', 'sugeno fuzzy logic'])\n", "('NONE', 'integrated model is developed by combining XXXXX ( ga ) , XXXXX ( rbf-nn ) and sugeno fuzzy logic approaches', ['genetic algorithm', 'radial basis function neural network'])\n", "('method used for task', 'integrated model is developed by combining XXXXX ( ga ) , radial basis function neural network ( rbf-nn ) and XXXXX approaches', ['genetic algorithm', 'sugeno fuzzy logic'])\n", "('method used for task', 'integrated model is developed by combining genetic algorithm ( ga ) , XXXXX ( rbf-nn ) and XXXXX approaches', ['radial basis function neural network', 'sugeno fuzzy logic'])\n", "('NONE', \"enhanced pid parameters are used to design sugeno XXXXX tuned by XXXXX ( ke , τe ) of the avr system to improve the system 's response\", ['fuzzy pid controller', 'excitation parameter'])\n", "('NONE', \"enhanced pid parameters are used to design sugeno XXXXX tuned by excitation parameter ( ke , τe ) of the XXXXX to improve the system 's response\", ['fuzzy pid controller', 'avr system'])\n", "('NONE', \"enhanced pid parameters are used to design sugeno fuzzy pid controller tuned by XXXXX ( ke , τe ) of the XXXXX to improve the system 's response\", ['excitation parameter', 'avr system'])\n", "('method used for task', 'we proposed a new multiobjective XXXXX based methodology for adaption of neural network structure for XXXXX of satellite imagery', ['particle swarm optimization', 'pixel classification'])\n", "('method used for task', 'we define a intuitionistic fuzzy parameterized XXXXX for dealing with uncertainties that is based on both XXXXX and XXXXX', ['soft sets', 'intuitionistic fuzzy sets'])\n", "('method used for task', 'we define a intuitionistic fuzzy parameterized XXXXX for dealing with uncertainties that is based on both XXXXX and XXXXX', ['soft sets', 'intuitionistic fuzzy sets'])\n", "('method used for task', 'the fuzziness of the XXXXX is analyzed in two different ways such as – triangular and XXXXX', ['trapezoidal fuzzy numbers', 'credit period'])\n", "('NONE', 'reformulation of the involved XXXXX in terms of penalized XXXXX for the pso implementation', ['constrained optimization problem', 'unconstrained optimization problem'])\n", "('method used for task', 'we have experimentally validated the optimization capabilities of these schemes and compared them to XXXXX , dynamic mesh optimization , and XXXXX', ['particle swarm optimization', 'firefly algorithm'])\n", "('NONE', 'an XXXXX that segments and classifies four XXXXX', ['integrated system', 'moving objects'])\n", "('method used for task', 'automatic XXXXX for aann and XXXXX and extraction methods are developed', ['data generation', 'data selection'])\n", "('method used for task', 'automatic XXXXX for aann and data selection and XXXXX are developed', ['data generation', 'extraction methods'])\n", "('method used for task', 'automatic data generation for aann and XXXXX and XXXXX are developed', ['data selection', 'extraction methods'])\n", "('method used for task', 'an estimation model of XXXXX based on XXXXX', ['missing data', 'multilayer perceptron'])\n", "('NONE', 'an XXXXX of XXXXX based on multilayer perceptron', ['missing data', 'estimation model'])\n", "('NONE', 'an XXXXX of missing data based on XXXXX', ['multilayer perceptron', 'estimation model'])\n", "('method used for task', 'combination of XXXXX and k-nearest neighbour-based XXXXX', ['neural network', 'multiple imputation'])\n", "('NONE', 'the optimization of the antecedent parameters for a type 2 XXXXX of XXXXX is presented', ['fuzzy system', 'edge detection'])\n", "('NONE', 'the goal of XXXXX in XXXXX is to provide the ability to handle uncertainty', ['interval type-2 fuzzy logic', 'edge detection methods'])\n", "('NONE', 'results show that the XXXXX provides better results in optimizing the type-2 XXXXX', ['cuckoo search', 'fuzzy system'])\n", "('method used for task', 'the main disadvantage of fwnn is that the application domain is limited to static problems due to its feed-forward XXXXX . therefore , we propose to use a self-recurrent XXXXX ( srwnn ) in the consequent part of fwnn , solving the control problem for chaotic systems', ['network structure', 'wavelet neural network'])\n", "('method used for task', 'the main disadvantage of fwnn is that the application domain is limited to static problems due to its feed-forward XXXXX . therefore , we propose to use a self-recurrent wavelet neural network ( srwnn ) in the consequent part of fwnn , solving the control problem for XXXXX', ['network structure', 'chaotic systems'])\n", "('NONE', 'the main disadvantage of fwnn is that the application domain is limited to static problems due to its feed-forward network structure . therefore , we propose to use a self-recurrent XXXXX ( srwnn ) in the consequent part of fwnn , solving the control problem for XXXXX', ['wavelet neural network', 'chaotic systems'])\n", "('NONE', 'finding the optimal learning rates is a challenging task in the classic gradient-based XXXXX . hence , in our proposed framework , all of the learning rates are determined optimally based on XXXXX', ['learning algorithms', 'lyapunov stability theory'])\n", "('method used for task', 'finding the optimal XXXXX is a challenging task in the classic gradient-based XXXXX . hence , in our proposed framework , all of the XXXXX are determined optimally based on lyapunov stability theory', ['learning algorithms', 'learning rates'])\n", "('method used for task', 'finding the optimal XXXXX is a challenging task in the classic gradient-based XXXXX . hence , in our proposed framework , all of the XXXXX are determined optimally based on lyapunov stability theory', ['learning algorithms', 'learning rates'])\n", "('NONE', 'finding the optimal XXXXX is a challenging task in the classic gradient-based learning algorithms . hence , in our proposed framework , all of the XXXXX are determined optimally based on XXXXX', ['lyapunov stability theory', 'learning rates'])\n", "('NONE', 'finding the optimal XXXXX is a challenging task in the classic gradient-based learning algorithms . hence , in our proposed framework , all of the XXXXX are determined optimally based on XXXXX', ['lyapunov stability theory', 'learning rates'])\n", "('method used for task', 'the merits of proposed methodology are XXXXX and less XXXXX', ['high accuracy', 'execution time'])\n", "('NONE', 'having the best performance of np-pso in solving various XXXXX compared with some well-known XXXXX', ['nonlinear functions', 'pso algorithms'])\n", "('NONE', 'XXXXX ( rbf ) XXXXX is developed on fpga', ['radial basis function', 'neural network'])\n", "('method used for task', 'XXXXX on randomly generated graphs and XXXXX of results', ['computational experiments', 'statistical analysis'])\n", "('NONE', 'the joint statistics and XXXXX of the pdtdfb XXXXX are studied', ['mutual information', 'transform coefficients'])\n", "('NONE', 'the pdtdfb transform coefficients are modeled using a hmt XXXXX with XXXXX', ['statistical model', 'gaussian mixtures'])\n", "('NONE', 'the pdtdfb XXXXX are modeled using a hmt XXXXX with gaussian mixtures', ['statistical model', 'transform coefficients'])\n", "('NONE', 'the pdtdfb XXXXX are modeled using a hmt statistical model with XXXXX', ['gaussian mixtures', 'transform coefficients'])\n", "('NONE', 'interaction of XXXXX with sbst combines XXXXX with experience', ['computational power', 'domain experts'])\n", "('NONE', 'the length of XXXXX of XXXXX could be adjusted adaptively to balance the computing time and solving ability', ['scatter search', 'reference set'])\n", "('method used for task', 'the length of reference set of XXXXX could be adjusted adaptively to balance the XXXXX and solving ability', ['scatter search', 'computing time'])\n", "('NONE', 'the length of XXXXX of scatter search could be adjusted adaptively to balance the XXXXX and solving ability', ['reference set', 'computing time'])\n", "('method used for task', 'based on karnik-mendel ( km ) algorithm , we propose an XXXXX to XXXXX , and apply it in personnel selection problems for knowledge-intensive enterprise', ['analytical solution', 'fuzzy topsis method'])\n", "('method used for task', 'orpd problem is formulated for real XXXXX and XXXXX minimization', ['power loss', 'voltage deviation'])\n", "('method used for task', 'soft computing techniques used : ontologies , XXXXX , and a XXXXX', ['statistical tests', 'classification algorithm'])\n", "('NONE', 'XXXXX used : ontologies , XXXXX , and a classification algorithm', ['statistical tests', 'soft computing techniques'])\n", "('method used for task', 'XXXXX used : ontologies , statistical tests , and a XXXXX', ['classification algorithm', 'soft computing techniques'])\n", "('method used for task', 'the new XXXXX without insignificant components are established through the XXXXX', ['regression models', 'stepwise regression'])\n", "('method used for task', 'adaptively computes the XXXXX for XXXXX and restoration purposes', ['fuzzy membership functions', 'noise detection'])\n", "('method used for task', 'selection of XXXXX and controller structure is vital for XXXXX', ['objective function', 'controller design'])\n", "('method used for task', 'selection of XXXXX and XXXXX is vital for controller design', ['objective function', 'controller structure'])\n", "('method used for task', 'selection of objective function and XXXXX is vital for XXXXX', ['controller design', 'controller structure'])\n", "('NONE', 'an XXXXX using itae , XXXXX and settling times is proposed', ['objective function', 'damping ratio'])\n", "('method used for task', 'apply XXXXX to dynamic XXXXX', ['hopfield neural network', 'parameter estimation'])\n", "('NONE', 'obtain the initial XXXXX of regular XXXXX', ['fuzzy classification', 'feature space'])\n", "('NONE', 'the objective is to reach standard levels of XXXXX with applying XXXXX of passive filters', ['harmonic distortion', 'minimum cost'])\n", "('method used for task', 'publicly available XXXXX were used for a XXXXX', ['environmental data', 'comparative study'])\n", "('method used for task', 'seven important XXXXX found to exist in XXXXX 50–300hz . a new method of selecting cwt scales according to the XXXXX found', ['fault frequencies', 'frequency range'])\n", "('method used for task', 'seven important XXXXX found to exist in XXXXX 50–300hz . a new method of selecting cwt scales according to the XXXXX found', ['frequency range', 'fault frequencies'])\n", "('method used for task', 'the mofca aims to balance the XXXXX over network by wisely adjusting the cluster competition radius according to three XXXXX', ['energy consumption', 'input parameters'])\n", "('method used for task', 'the XXXXX is based on XXXXX', ['quality evaluation', 'fuzzy classification'])\n", "('method used for task', 'an XXXXX and a heuristic are presented to solve the XXXXX', ['integer programming', 'np-hard problem'])\n", "('NONE', 'proposing a five- tiers XXXXX with flexible XXXXX', ['supply chain', 'lead time'])\n", "('method used for task', 'XXXXX is indispensable when dealing with XXXXX', ['feature selection', 'microarray data'])\n", "('method used for task', 'combination of pnn , XXXXX and XXXXX', ['hierarchical clustering', 'cluster validity index'])\n", "('method used for task', 'multiobjective spatial XXXXX for XXXXX is proposed', ['fuzzy clustering', 'image segmentation'])\n", "('NONE', 'a premise synchronizer has been delicately constructed to ensure the same premises with uniform XXXXX in both XXXXX and fuzzy controller', ['time scales', 'fuzzy model'])\n", "('NONE', 'a premise synchronizer has been delicately constructed to ensure the same premises with uniform XXXXX in both fuzzy model and XXXXX', ['time scales', 'fuzzy controller'])\n", "('method used for task', 'a premise synchronizer has been delicately constructed to ensure the same premises with uniform time scales in both XXXXX and XXXXX', ['fuzzy model', 'fuzzy controller'])\n", "('method used for task', 'XXXXX are identified by fmea and XXXXX', ['critical parameters', 'fuzzy fmea'])\n", "('NONE', 'we propose an improved XXXXX which makes the best of ergodicity of piecewise linear XXXXX to explore the global search while utilizing the sequential quadratic programming to accelerate the local search', ['gravitational search algorithm', 'chaotic map'])\n", "('NONE', 'we propose an improved XXXXX which makes the best of ergodicity of piecewise linear chaotic map to explore the XXXXX while utilizing the sequential quadratic programming to accelerate the local search', ['gravitational search algorithm', 'global search'])\n", "('NONE', 'we propose an improved XXXXX which makes the best of ergodicity of piecewise linear chaotic map to explore the global search while utilizing the sequential quadratic programming to accelerate the XXXXX', ['gravitational search algorithm', 'local search'])\n", "('method used for task', 'we propose an improved gravitational search algorithm which makes the best of ergodicity of piecewise linear XXXXX to explore the XXXXX while utilizing the sequential quadratic programming to accelerate the local search', ['chaotic map', 'global search'])\n", "('method used for task', 'we propose an improved gravitational search algorithm which makes the best of ergodicity of piecewise linear XXXXX to explore the global search while utilizing the sequential quadratic programming to accelerate the XXXXX', ['chaotic map', 'local search'])\n", "('method used for task', 'we propose an improved gravitational search algorithm which makes the best of ergodicity of piecewise linear chaotic map to explore the XXXXX while utilizing the sequential quadratic programming to accelerate the XXXXX', ['global search', 'local search'])\n", "('method used for task', 'based on the binary improved XXXXX and the XXXXX ( k-nn ) method , a novel hybrid system is proposed to improve classification accuracy with an appropriate feature subset in binary problems', ['gravitational search algorithm', 'k-nearest neighbor'])\n", "('method used for task', 'based on the binary improved XXXXX and the k-nearest neighbor ( k-nn ) method , a novel XXXXX is proposed to improve classification accuracy with an appropriate feature subset in binary problems', ['gravitational search algorithm', 'hybrid system'])\n", "('method used for task', 'based on the binary improved XXXXX and the k-nearest neighbor ( k-nn ) method , a novel hybrid system is proposed to improve XXXXX with an appropriate feature subset in binary problems', ['gravitational search algorithm', 'classification accuracy'])\n", "('NONE', 'based on the binary improved gravitational search algorithm and the XXXXX ( k-nn ) method , a novel XXXXX is proposed to improve classification accuracy with an appropriate feature subset in binary problems', ['k-nearest neighbor', 'hybrid system'])\n", "('method used for task', 'based on the binary improved gravitational search algorithm and the XXXXX ( k-nn ) method , a novel hybrid system is proposed to improve XXXXX with an appropriate feature subset in binary problems', ['k-nearest neighbor', 'classification accuracy'])\n", "('method used for task', 'based on the binary improved gravitational search algorithm and the k-nearest neighbor ( k-nn ) method , a novel XXXXX is proposed to improve XXXXX with an appropriate feature subset in binary problems', ['hybrid system', 'classification accuracy'])\n", "('NONE', 'three XXXXX based XXXXX such as , ann–ga , ann–sa and ann–quasi newton have been developed', ['soft computing', 'integrated models'])\n", "('NONE', 'XXXXX on four different XXXXX ( non-neutral and neutral nk landscapes , flow-shop , qap , maxsat )', ['experimental analysis', 'landscape models'])\n", "('method used for task', 'm-band XXXXX is used to extract scale-space features for XXXXX', ['wavelet packet', 'document image'])\n", "('method used for task', 'er-wca outperforms or equals other methods in terms of XXXXX and XXXXX', ['function evaluations', 'solution quality'])\n", "('NONE', 'human visual perception in assessing XXXXX using type 2 XXXXX', ['image quality', 'fuzzy sets'])\n", "('method used for task', 'we design an XXXXX which is suitable for process with XXXXX', ['control chart', 'fuzzy parameters'])\n", "('NONE', 'we solve the XXXXX using the combination of two XXXXX', ['evolutionary methods', 'path planning problem'])\n", "('NONE', 'second , XXXXX ( ep ) optimizes the XXXXX and smoothness', ['evolutionary programming', 'path length'])\n", "('method used for task', 'also developed hybrid of XXXXX and adaptive XXXXX', ['score level', 'fuzzy decision level'])\n", "('NONE', 'the fmcdm model avoids multiplying two XXXXX into a pooled XXXXX , and reserves fuzzy information', ['fuzzy numbers', 'fuzzy number'])\n", "('NONE', 'the effect of degree of XXXXX and hybrid combination of XXXXX is studied', ['activation functions', 'input variables'])\n", "('method used for task', 'we propose a directional XXXXX for XXXXX ( de )', ['mutation operator', 'differential evolution'])\n", "('NONE', 'this XXXXX the permutation flowshop scheduling problem with XXXXX and weighted tardiness', ['order acceptance', 'paper studies'])\n", "('NONE', 'this paper studies the XXXXX with XXXXX and weighted tardiness', ['order acceptance', 'permutation flowshop scheduling problem'])\n", "('NONE', 'this XXXXX the XXXXX with order acceptance and weighted tardiness', ['paper studies', 'permutation flowshop scheduling problem'])\n", "('NONE', 'nearest XXXXX of continuous type-2 fuzzy variable with cv-based XXXXX', ['interval approximation', 'reduction method'])\n", "('NONE', 'a new XXXXX , combining XXXXX and adaptive-network-based fuzzy inference system ( anfis–pso ) was used', ['hybrid approach', 'particle swarm optimization'])\n", "('method used for task', 'a new XXXXX , combining particle swarm optimization and adaptive-network-based XXXXX ( anfis–pso ) was used', ['hybrid approach', 'fuzzy inference system'])\n", "('method used for task', 'a new hybrid approach , combining XXXXX and adaptive-network-based XXXXX ( anfis–pso ) was used', ['particle swarm optimization', 'fuzzy inference system'])\n", "('method used for task', 'a thresholding method is proposed using XXXXX and XXXXX', ['fuzzy entropy', 'bat algorithm'])\n", "('NONE', 'a XXXXX is proposed using XXXXX and bat algorithm', ['fuzzy entropy', 'thresholding method'])\n", "('method used for task', 'a XXXXX is proposed using fuzzy entropy and XXXXX', ['bat algorithm', 'thresholding method'])\n", "('NONE', 'the proposed method can stably converge to XXXXX with XXXXX', ['optimal threshold', 'high efficiency'])\n", "('method used for task', 'we present a bi-objective XXXXX for a XXXXX', ['inventory model', 'supply chain problem'])\n", "('method used for task', 'it2fls employs hybrid learning process by XXXXX and XXXXX', ['fuzzy c-means', 'genetic algorithm'])\n", "('NONE', 'to evaluate the gsd team-level service climate and gsd project outcome relationship based on adaptive neuro-fuzzy XXXXX ( anfis ) with the hybrid taguchi-genetic XXXXX ( htgla )', ['inference system', 'learning algorithm'])\n", "('method used for task', 'to evaluate the gsd team-level XXXXX and gsd project outcome relationship based on adaptive neuro-fuzzy XXXXX ( anfis ) with the hybrid taguchi-genetic learning algorithm ( htgla )', ['inference system', 'service climate'])\n", "('method used for task', 'to evaluate the gsd team-level XXXXX and gsd project outcome relationship based on adaptive neuro-fuzzy inference system ( anfis ) with the hybrid taguchi-genetic XXXXX ( htgla )', ['learning algorithm', 'service climate'])\n", "('NONE', 'our evolutionary XXXXX provides improved XXXXX', ['recognition rate', 'face recognition algorithm'])\n", "('NONE', 'we obtain the XXXXX of XXXXX induced by soft coverings', ['lattice structure', 'soft sets'])\n", "('method used for task', 'we investigate the parameter reduction of soft coverings by means of knowledge on the XXXXX for covering XXXXX', ['attribute reduction', 'information systems'])\n", "('NONE', 'we investigate the XXXXX of soft coverings by means of knowledge on the XXXXX for covering information systems', ['attribute reduction', 'parameter reduction'])\n", "('NONE', 'we investigate the XXXXX of soft coverings by means of knowledge on the attribute reduction for covering XXXXX', ['information systems', 'parameter reduction'])\n", "('NONE', 'the development of a XXXXX capable of determining the number of daily deals present on a given XXXXX , and accurately localizing those segments of the page that consist of information about one particular deal', ['segmentation algorithm', 'web page'])\n", "('method used for task', 'we propose XXXXX and XXXXX to solve the problem', ['genetic algorithm', 'particle swarm optimization'])\n", "('NONE', 'to improve XXXXX , aco based XXXXX for pmc is proposed', ['classification accuracy', 'feature selection'])\n", "('NONE', 'presenting an efficient XXXXX by combining XXXXX and relaxation induced neighborhood search to solve the proposed model', ['hybrid algorithm', 'local branching'])\n", "('method used for task', 'presenting an efficient XXXXX by combining local branching and relaxation induced XXXXX to solve the proposed model', ['hybrid algorithm', 'neighborhood search'])\n", "('method used for task', 'presenting an efficient hybrid algorithm by combining XXXXX and relaxation induced XXXXX to solve the proposed model', ['local branching', 'neighborhood search'])\n", "('NONE', 'the object is divided into m overlapped patches to improve XXXXX in XXXXX and partial occlusions', ['tracking performance', 'background clutter'])\n", "('NONE', 'during a XXXXX , a XXXXX is constructed to minimize the total probability of misclassifying the labeled training examples , a process which approximately maximizes the margins of the resulting classifier', ['decision tree', 'training phase'])\n", "('method used for task', 'we propose an adaptive takagi-sugeno-kang-fuzzy ( tsk-fuzzy ) XXXXX ( atfsc ) for use in direct torque control ( dtc ) induction motor ( im ) drives to improve their XXXXX', ['dynamic responses', 'speed controller'])\n", "('method used for task', 'we propose an adaptive takagi-sugeno-kang-fuzzy ( tsk-fuzzy ) speed controller ( atfsc ) for use in direct XXXXX ( dtc ) induction motor ( im ) drives to improve their XXXXX', ['dynamic responses', 'torque control'])\n", "('method used for task', 'we propose an adaptive takagi-sugeno-kang-fuzzy ( tsk-fuzzy ) XXXXX ( atfsc ) for use in direct XXXXX ( dtc ) induction motor ( im ) drives to improve their dynamic responses', ['speed controller', 'torque control'])\n", "('NONE', 'the atfsc , XXXXX , and pi control schemes were experimentally investigated , using the root mean square error ( rmse ) XXXXX to evaluate each scheme', ['fuzzy control', 'performance index'])\n", "('NONE', 'the atfsc , XXXXX , and pi control schemes were experimentally investigated , using the XXXXX square error ( rmse ) performance index to evaluate each scheme', ['fuzzy control', 'root mean'])\n", "('NONE', 'the atfsc , fuzzy control , and pi control schemes were experimentally investigated , using the XXXXX square error ( rmse ) XXXXX to evaluate each scheme', ['performance index', 'root mean'])\n", "('method used for task', 'the system uses : XXXXX and XXXXX to self-adjust the foa', ['genetic fuzzy systems', 'genetic programming'])\n", "('method used for task', 'this paper proposes an improved multi-objective XXXXX chemotaxis algorithm to solve XXXXX', ['bacteria colony', 'multi-objective optimization problems'])\n", "('method used for task', 'the XXXXX obtained by the improved algorithm has better distribution and convergence than other XXXXX', ['pareto front', 'optimization algorithms'])\n", "('NONE', 'the XXXXX of the general pareto-based multi-objective XXXXX chemotaxis algorithm is proved', ['convergence property', 'bacteria colony'])\n", "('method used for task', 'applications were done on some XXXXX and real-world datasets taken from uci XXXXX', ['benchmark data', 'machine learning repository'])\n", "('NONE', 'XXXXX are number of solar panels , XXXXX and batteries', ['wind turbines', 'decision variables'])\n", "('NONE', 'XXXXX ( de ) , XXXXX ( pso ) and game-theoretic XXXXX ( gtde ) algorithms are implemented', ['differential evolution', 'particle swarm optimization'])\n", "('NONE', 'XXXXX ( de ) , XXXXX ( pso ) and game-theoretic XXXXX ( gtde ) algorithms are implemented', ['particle swarm optimization', 'differential evolution'])\n", "('NONE', 'XXXXX provide adaptive XXXXX for public health protection', ['dynamic models', 'decision support'])\n", "('NONE', 'we presented new hardware architectures for XXXXX of several it2 flcs ( km , wm , bmm , nt ) with the consideration of limitations : device area and XXXXX', ['hardware implementation', 'real-time performance'])\n", "('method used for task', 'we presented new XXXXX for XXXXX of several it2 flcs ( km , wm , bmm , nt ) with the consideration of limitations : device area and real-time performance', ['hardware implementation', 'hardware architectures'])\n", "('NONE', 'we presented new XXXXX for hardware implementation of several it2 flcs ( km , wm , bmm , nt ) with the consideration of limitations : device area and XXXXX', ['real-time performance', 'hardware architectures'])\n", "('method used for task', 'we apply ga and XXXXX to find XXXXX for the entire network', ['pso algorithm', 'optimal solution'])\n", "('NONE', 'a bidirectional XXXXX r-variable XXXXX ( biorv-nsa ) is proposed to generate less mature detectors and cover more “black holes”', ['negative selection algorithm', 'inhibition optimization'])\n", "('NONE', 'when doing comparative experiments , biorv-nsa is divided into two kinds of algorithms , which are a bidirectional weak XXXXX r-variable XXXXX and a bidirectional strong XXXXX r-variable XXXXX , respectively', ['negative selection algorithm', 'inhibition optimization'])\n", "('NONE', 'when doing comparative experiments , biorv-nsa is divided into two kinds of algorithms , which are a bidirectional weak XXXXX r-variable XXXXX and a bidirectional strong XXXXX r-variable XXXXX , respectively', ['negative selection algorithm', 'inhibition optimization'])\n", "('NONE', 'when doing comparative experiments , biorv-nsa is divided into two kinds of algorithms , which are a bidirectional weak XXXXX r-variable XXXXX and a bidirectional strong XXXXX r-variable XXXXX , respectively', ['negative selection algorithm', 'inhibition optimization'])\n", "('NONE', 'when doing comparative experiments , biorv-nsa is divided into two kinds of algorithms , which are a bidirectional weak XXXXX r-variable XXXXX and a bidirectional strong XXXXX r-variable XXXXX , respectively', ['negative selection algorithm', 'inhibition optimization'])\n", "('method used for task', 'the objective of this novel XXXXX is to increase the XXXXX for data set with outliers', ['support vector regression', 'generalization ability'])\n", "('method used for task', 'a new XXXXX for power system disturbances is introduced using XXXXX', ['learning method', 'extreme learning machine'])\n", "('method used for task', 'simultaneously optimize the XXXXX and XXXXX for the elm using pso', ['model selection', 'feature subset'])\n", "('method used for task', 'proposed method can improve XXXXX and XXXXX of elm', ['generalization performance', 'convergence accuracy'])\n", "('NONE', 'XXXXX using XXXXX', ['principal component analysis', 'feature dimension reduction'])\n", "('NONE', 'XXXXX was built via XXXXX ( ann )', ['prediction model', 'artificial neural network'])\n", "('NONE', 'pe has better XXXXX in XXXXX', ['generalization ability', 'activity recognition'])\n", "('NONE', 'the study models the problem by bi-level XXXXX . this study develops a revised ant algorithm with efficient XXXXX to solve the problem', ['stochastic programming', 'greedy heuristics'])\n", "('method used for task', 'a XXXXX is proposed for efficient XXXXX in high-dimensional datasets . the symmetrical uncertainty ( su ) criterion is exploited to weight features in filter phase', ['hybrid method', 'subset selection'])\n", "('NONE', 'in wrapper phase , both fuzzy imperialist competitive algorithm ( fica ) and incremental wrapper XXXXX with replacement ( iwssr ) in weighted XXXXX are executed to search and find relevant attributes', ['subset selection', 'feature space'])\n", "('NONE', '31 components of XXXXX are tested with a systematic XXXXX', ['abc algorithm', 'experimental study'])\n", "('method used for task', 'two XXXXX , soco and cec05 , are used in XXXXX', ['experimental study', 'benchmark sets'])\n", "('NONE', 'two new variants of XXXXX are proposed for each of the two XXXXX', ['abc algorithms', 'benchmark sets'])\n", "('NONE', 'proposed methodology uses the current signal in XXXXX as the inputs of the pattern classifiers for XXXXX', ['time domain', 'fault diagnosis'])\n", "('NONE', 'we propose a hybrid collaborative XXXXX on fuzzy social-only XXXXX and local search methods for dynamic optimization problems', ['model based', 'particle swarm optimization'])\n", "('method used for task', 'we propose a hybrid collaborative XXXXX on fuzzy social-only particle swarm optimization and XXXXX for dynamic optimization problems', ['model based', 'local search methods'])\n", "('method used for task', 'we propose a hybrid collaborative model based on fuzzy social-only XXXXX and XXXXX for dynamic optimization problems', ['particle swarm optimization', 'local search methods'])\n", "('NONE', 'the XXXXX , in precise XXXXX ( rst ) based class information and edge information is used in bilateral framework for medical image denoising problem', ['soft computing', 'rough set theory'])\n", "('method used for task', 'the XXXXX , in precise rough set theory ( rst ) based class information and XXXXX is used in bilateral framework for medical image denoising problem', ['soft computing', 'edge information'])\n", "('NONE', 'the XXXXX , in precise rough set theory ( rst ) based XXXXX and edge information is used in bilateral framework for medical image denoising problem', ['soft computing', 'class information'])\n", "('method used for task', 'the soft computing , in precise XXXXX ( rst ) based class information and XXXXX is used in bilateral framework for medical image denoising problem', ['rough set theory', 'edge information'])\n", "('NONE', 'the soft computing , in precise XXXXX ( rst ) based XXXXX and edge information is used in bilateral framework for medical image denoising problem', ['rough set theory', 'class information'])\n", "('method used for task', 'the soft computing , in precise rough set theory ( rst ) based XXXXX and XXXXX is used in bilateral framework for medical image denoising problem', ['edge information', 'class information'])\n", "('method used for task', 'as the XXXXX , cart ( classification and XXXXX ) , lsr ( least squares regression ) , glr ( generalized linear regression ) , mvlr ( multivariate linear regression ) , pslr ( partial least squares regression ) , grnn ( generalized regression neural network ) , mlp ( multilayer perceptron ) and svr ( support vector regression ) are used', ['machine learning algorithms', 'regression trees'])\n", "('method used for task', 'as the XXXXX , cart ( classification and regression trees ) , lsr ( least squares regression ) , glr ( generalized XXXXX ) , mvlr ( multivariate XXXXX ) , pslr ( partial least squares regression ) , grnn ( generalized regression neural network ) , mlp ( multilayer perceptron ) and svr ( support vector regression ) are used', ['machine learning algorithms', 'linear regression'])\n", "('method used for task', 'as the XXXXX , cart ( classification and regression trees ) , lsr ( least squares regression ) , glr ( generalized XXXXX ) , mvlr ( multivariate XXXXX ) , pslr ( partial least squares regression ) , grnn ( generalized regression neural network ) , mlp ( multilayer perceptron ) and svr ( support vector regression ) are used', ['machine learning algorithms', 'linear regression'])\n", "('method used for task', 'as the XXXXX , cart ( classification and regression trees ) , lsr ( least squares regression ) , glr ( generalized linear regression ) , mvlr ( multivariate linear regression ) , pslr ( partial least squares regression ) , grnn ( generalized regression neural network ) , mlp ( XXXXX ) and svr ( support vector regression ) are used', ['machine learning algorithms', 'multilayer perceptron'])\n", "('method used for task', 'as the XXXXX , cart ( classification and regression trees ) , lsr ( least squares regression ) , glr ( generalized linear regression ) , mvlr ( multivariate linear regression ) , pslr ( partial least squares regression ) , grnn ( generalized regression neural network ) , mlp ( multilayer perceptron ) and svr ( XXXXX ) are used', ['machine learning algorithms', 'support vector regression'])\n", "('method used for task', 'as the XXXXX , cart ( classification and regression trees ) , lsr ( least squares regression ) , glr ( generalized linear regression ) , mvlr ( multivariate linear regression ) , pslr ( partial least squares regression ) , grnn ( generalized XXXXX ) , mlp ( multilayer perceptron ) and svr ( support vector regression ) are used', ['machine learning algorithms', 'regression neural network'])\n", "('NONE', 'as the machine learning algorithms , cart ( classification and XXXXX ) , lsr ( least squares regression ) , glr ( generalized XXXXX ) , mvlr ( multivariate XXXXX ) , pslr ( partial least squares regression ) , grnn ( generalized regression neural network ) , mlp ( multilayer perceptron ) and svr ( support vector regression ) are used', ['regression trees', 'linear regression'])\n", "('NONE', 'as the machine learning algorithms , cart ( classification and XXXXX ) , lsr ( least squares regression ) , glr ( generalized XXXXX ) , mvlr ( multivariate XXXXX ) , pslr ( partial least squares regression ) , grnn ( generalized regression neural network ) , mlp ( multilayer perceptron ) and svr ( support vector regression ) are used', ['regression trees', 'linear regression'])\n", "('NONE', 'as the machine learning algorithms , cart ( classification and XXXXX ) , lsr ( least squares regression ) , glr ( generalized linear regression ) , mvlr ( multivariate linear regression ) , pslr ( partial least squares regression ) , grnn ( generalized regression neural network ) , mlp ( XXXXX ) and svr ( support vector regression ) are used', ['regression trees', 'multilayer perceptron'])\n", "('method used for task', 'as the machine learning algorithms , cart ( classification and XXXXX ) , lsr ( least squares regression ) , glr ( generalized linear regression ) , mvlr ( multivariate linear regression ) , pslr ( partial least squares regression ) , grnn ( generalized regression neural network ) , mlp ( multilayer perceptron ) and svr ( XXXXX ) are used', ['regression trees', 'support vector regression'])\n", "('NONE', 'as the machine learning algorithms , cart ( classification and XXXXX ) , lsr ( least squares regression ) , glr ( generalized linear regression ) , mvlr ( multivariate linear regression ) , pslr ( partial least squares regression ) , grnn ( generalized XXXXX ) , mlp ( multilayer perceptron ) and svr ( support vector regression ) are used', ['regression trees', 'regression neural network'])\n", "('NONE', 'as the machine learning algorithms , cart ( classification and regression trees ) , lsr ( least squares regression ) , glr ( generalized XXXXX ) , mvlr ( multivariate XXXXX ) , pslr ( partial least squares regression ) , grnn ( generalized regression neural network ) , mlp ( XXXXX ) and svr ( support vector regression ) are used', ['linear regression', 'multilayer perceptron'])\n", "('NONE', 'as the machine learning algorithms , cart ( classification and regression trees ) , lsr ( least squares regression ) , glr ( generalized XXXXX ) , mvlr ( multivariate XXXXX ) , pslr ( partial least squares regression ) , grnn ( generalized regression neural network ) , mlp ( multilayer perceptron ) and svr ( XXXXX ) are used', ['linear regression', 'support vector regression'])\n", "('NONE', 'as the machine learning algorithms , cart ( classification and regression trees ) , lsr ( least squares regression ) , glr ( generalized XXXXX ) , mvlr ( multivariate XXXXX ) , pslr ( partial least squares regression ) , grnn ( generalized XXXXX ) , mlp ( multilayer perceptron ) and svr ( support vector regression ) are used', ['linear regression', 'regression neural network'])\n", "('NONE', 'as the machine learning algorithms , cart ( classification and regression trees ) , lsr ( least squares regression ) , glr ( generalized XXXXX ) , mvlr ( multivariate XXXXX ) , pslr ( partial least squares regression ) , grnn ( generalized regression neural network ) , mlp ( XXXXX ) and svr ( support vector regression ) are used', ['linear regression', 'multilayer perceptron'])\n", "('NONE', 'as the machine learning algorithms , cart ( classification and regression trees ) , lsr ( least squares regression ) , glr ( generalized XXXXX ) , mvlr ( multivariate XXXXX ) , pslr ( partial least squares regression ) , grnn ( generalized regression neural network ) , mlp ( multilayer perceptron ) and svr ( XXXXX ) are used', ['linear regression', 'support vector regression'])\n", "('NONE', 'as the machine learning algorithms , cart ( classification and regression trees ) , lsr ( least squares regression ) , glr ( generalized XXXXX ) , mvlr ( multivariate XXXXX ) , pslr ( partial least squares regression ) , grnn ( generalized XXXXX ) , mlp ( multilayer perceptron ) and svr ( support vector regression ) are used', ['linear regression', 'regression neural network'])\n", "('method used for task', 'as the machine learning algorithms , cart ( classification and regression trees ) , lsr ( least squares regression ) , glr ( generalized linear regression ) , mvlr ( multivariate linear regression ) , pslr ( partial least squares regression ) , grnn ( generalized regression neural network ) , mlp ( XXXXX ) and svr ( XXXXX ) are used', ['multilayer perceptron', 'support vector regression'])\n", "('NONE', 'as the machine learning algorithms , cart ( classification and regression trees ) , lsr ( least squares regression ) , glr ( generalized linear regression ) , mvlr ( multivariate linear regression ) , pslr ( partial least squares regression ) , grnn ( generalized XXXXX ) , mlp ( XXXXX ) and svr ( support vector regression ) are used', ['multilayer perceptron', 'regression neural network'])\n", "('method used for task', 'as the machine learning algorithms , cart ( classification and regression trees ) , lsr ( least squares regression ) , glr ( generalized linear regression ) , mvlr ( multivariate linear regression ) , pslr ( partial least squares regression ) , grnn ( generalized XXXXX ) , mlp ( multilayer perceptron ) and svr ( XXXXX ) are used', ['support vector regression', 'regression neural network'])\n", "('method used for task', 'a surrogate based on XXXXX is adapted for mixed-type variables , XXXXX and constraints and integrated into nsga-ii', ['radial basis function networks', 'multiple objectives'])\n", "('method used for task', 'variants of nsga-ii including these changes are applied to a typical building optimisation problem , with improvements in XXXXX and XXXXX', ['convergence speed', 'solution quality'])\n", "('method used for task', 'proposing XXXXX to raise the capability for XXXXX and raise the system accuracy', ['fuzzy inference system', 'fault tolerance'])\n", "('NONE', 'the on-line adaptation mechanism of neurocontroller weights relies on levenberg–marquardt method with XXXXX for adaptive changes of a XXXXX', ['fuzzy model', 'learning rate'])\n", "('NONE', 'the developed speed neurocontroller uses only easy measurable XXXXX ( no sensors or estimators of other mechanical XXXXX are applied )', ['motor speed', 'state variables'])\n", "('method used for task', 'XXXXX optimized online anfis based speed controller presented for XXXXX', ['bat algorithm', 'brushless dc motor'])\n", "('method used for task', 'XXXXX optimized online anfis based XXXXX presented for brushless dc motor', ['bat algorithm', 'speed controller'])\n", "('method used for task', 'bat algorithm optimized online anfis based XXXXX presented for XXXXX', ['brushless dc motor', 'speed controller'])\n", "('NONE', 'a mixed XXXXX ( mip ) , a XXXXX and four meta-heuristics are developed', ['local search', 'integer model'])\n", "('method used for task', 'developing algorithms to extract training patterns from XXXXX for constructing XXXXX', ['historical data', 'forecasting models'])\n", "('method used for task', 'the ea-bnb achieved improvements regarding both XXXXX and XXXXX', ['cpu time', 'solution quality'])\n", "('NONE', 'we propose an asynchronous XXXXX of XXXXX ( de )', ['parallel implementation', 'differential evolution'])\n", "('NONE', 'tested with benchmarks problems , including the calibration of non-linear XXXXX of XXXXX', ['dynamic models', 'biological systems'])\n", "('method used for task', 'the XXXXX influence the population space to guide the feasible XXXXX', ['search space', 'knowledge sources'])\n", "('method used for task', 'contribution : hybridizing XXXXX and XXXXX', ['linear programming', 'genetic algorithm'])\n", "('method used for task', 'XXXXX is preserved by proposed XXXXX algorithm', ['image quality', 'pixel selection'])\n", "('NONE', 'XXXXX which include XXXXX , feature selection and classification for determining the disc abnormalities of the lumbar region have been developed', ['hybrid models', 'feature extraction'])\n", "('NONE', 'XXXXX which include feature extraction , XXXXX and classification for determining the disc abnormalities of the lumbar region have been developed', ['hybrid models', 'feature selection'])\n", "('method used for task', 'XXXXX which include feature extraction , feature selection and classification for determining the XXXXX of the lumbar region have been developed', ['hybrid models', 'disc abnormalities'])\n", "('NONE', 'hybrid models which include XXXXX , XXXXX and classification for determining the disc abnormalities of the lumbar region have been developed', ['feature extraction', 'feature selection'])\n", "('method used for task', 'hybrid models which include XXXXX , feature selection and classification for determining the XXXXX of the lumbar region have been developed', ['feature extraction', 'disc abnormalities'])\n", "('method used for task', 'hybrid models which include feature extraction , XXXXX and classification for determining the XXXXX of the lumbar region have been developed', ['feature selection', 'disc abnormalities'])\n", "('NONE', 'the results obtained suggest that the proposed XXXXX can be safely used in the detection of XXXXX', ['hybrid models', 'disc abnormalities'])\n", "('NONE', 'the model utilizes XXXXX patterns and socio-economic data in the XXXXX data', ['activity sequence', 'time use'])\n", "('NONE', 'the XXXXX of different XXXXX are examined', ['classification methods', 'classification performances'])\n", "('method used for task', 'we modify the XXXXX based on prefetching to fully exploit the potential map tasks with XXXXX in section 4.3.1. this method has the advantages of reducing network transmission . furthermore , we consider part of nodes , whose remaining time is less then threshold tunder to avoid invalid data prefetching', ['scheduling algorithm', 'data locality'])\n", "('NONE', 'age and sms-emoa typically achieve the XXXXX of the true XXXXX', ['best approximation', 'pareto front'])\n", "('method used for task', 'XXXXX : adjust parameters and XXXXX to the latest information', ['incremental learning', 'model structure'])\n", "('NONE', 'the proposed method is a XXXXX that combines XXXXX and k-means', ['hybrid method', 'the neural networks'])\n", "('method used for task', 'XXXXX with random small steps ( exploitation ) is performed only if XXXXX fails', ['local search', 'global search'])\n", "('method used for task', 'XXXXX to XXXXX ( odes ) in engineering', ['approximate solutions', 'ordinary differential equations'])\n", "('method used for task', 'we create a XXXXX to solve the XXXXX', ['modified differential evolution', 'optimization problem'])\n", "('method used for task', 'a new XXXXX is developed to reduce XXXXX for vehicle routing', ['mathematical model', 'co2 emissions'])\n", "('method used for task', 'a new XXXXX is developed to reduce co2 emissions for XXXXX', ['mathematical model', 'vehicle routing'])\n", "('method used for task', 'a new mathematical model is developed to reduce XXXXX for XXXXX', ['co2 emissions', 'vehicle routing'])\n", "('NONE', 'XXXXX can be reduced by taking traffic congestion into account in XXXXX', ['co2 emissions', 'vehicle routes'])\n", "('NONE', 'a soft computing method is firstly used in the literature to determine the XXXXX of tissue equivalent liquid exposed to XXXXX', ['temperature distribution', 'electromagnetic field'])\n", "('method used for task', 'a XXXXX is firstly used in the literature to determine the XXXXX of tissue equivalent liquid exposed to electromagnetic field', ['temperature distribution', 'soft computing method'])\n", "('method used for task', 'a XXXXX is firstly used in the literature to determine the temperature distribution of tissue equivalent liquid exposed to XXXXX', ['electromagnetic field', 'soft computing method'])\n", "('method used for task', 'a switching adaptive control scheme using a hopfield-based dynamic XXXXX ( sachnn ) for XXXXX with external disturbances is proposed', ['neural network', 'nonlinear systems'])\n", "('method used for task', 'we propose a model which incorporates kernel metric and XXXXX for XXXXX', ['fuzzy logic', 'image segmentation'])\n", "('NONE', 'nox and soot emissions from a XXXXX are modeled using XXXXX', ['diesel engine', 'artificial neural network model'])\n", "('NONE', 'the data used for training and testing the XXXXX are obtained through testing a XXXXX', ['diesel engine', 'ann model'])\n", "('NONE', 'the mass fuel rate , intake XXXXX , etc . are optimized using XXXXX', ['air temperature', 'ant colony optimization algorithm'])\n", "('NONE', 'a cheap and portable approach to detect XXXXX in XXXXX is proposed', ['fall detection', 'real time'])\n", "('method used for task', 'XXXXX are gathered by a wearable sensor and sent to a XXXXX', ['mobile device', 'acceleration data'])\n", "('NONE', 'provide an updated and XXXXX of distributed XXXXX', ['systematic review', 'evolutionary algorithms'])\n", "('NONE', 'location-routing XXXXX including XXXXX and direct shipment is modeled', ['scheduling problem', 'cross docking'])\n", "('method used for task', 'by the use of the bi-clustering method , XXXXX are solved by XXXXX', ['large-scale problems', 'exact methods'])\n", "('method used for task', 'XXXXX for modular robots equipped with XXXXX in complex environments', ['motion planning', 'motion primitives'])\n", "('method used for task', 'novel method for adaptation of the XXXXX based on XXXXX is proposed', ['particle swarm optimization', 'motion primitives'])\n", "('NONE', 'to reduce the impact of XXXXX , we incorporate XXXXX into antibody–antigen affinity', ['speckle noise', 'local information'])\n", "('method used for task', 'a new XXXXX is proposed for the jso that incorporates the XXXXX in the local search procedure', ['memetic algorithm', 'neighborhood structures'])\n", "('method used for task', 'a new XXXXX is proposed for the jso that incorporates the neighborhood structures in the XXXXX', ['memetic algorithm', 'local search procedure'])\n", "('NONE', 'a new memetic algorithm is proposed for the jso that incorporates the XXXXX in the XXXXX', ['neighborhood structures', 'local search procedure'])\n", "('NONE', 'an XXXXX was conducted showing that the XXXXX compares favorably with the state-of-the-art', ['experimental study', 'memetic algorithm'])\n", "('NONE', 'incorporation of XXXXX into XXXXX could be beneficial to text categorization task', ['topic identification', 'som learning'])\n", "('method used for task', 'two layered cascade-based XXXXX are proposed for XXXXX', ['evolutionary algorithms', 'feature selection'])\n", "('method used for task', 'both a XXXXX and an adaptive XXXXX are employed for expression recognition', ['neural network', 'ensemble classifier'])\n", "('method used for task', 'both a XXXXX and an adaptive ensemble classifier are employed for XXXXX', ['neural network', 'expression recognition'])\n", "('method used for task', 'both a neural network and an adaptive XXXXX are employed for XXXXX', ['ensemble classifier', 'expression recognition'])\n", "('method used for task', 'the orthogonal XXXXX is integrated into XXXXX', ['differential evolution', 'initialization method'])\n", "('method used for task', 'modified XXXXX is used to control XXXXX with simulated annealing technique', ['mutation operator', 'convergence rate'])\n", "('NONE', 'our XXXXX can only use XXXXX to mitigate the imbalance of the loads', ['distributed algorithm', 'local information'])\n", "('method used for task', 'a new color image segmentation method based on improved bee XXXXX for XXXXX ( ibmo ) is presented', ['multi-objective optimization', 'colony algorithm'])\n", "('NONE', 'the obtained results are compared the ones obtained from XXXXX which is one of the most popular methods used in XXXXX , nondominated sorting genetic algorithm , and nondominated sorted particle swarm optimization', ['fuzzy c-means', 'image segmentation'])\n", "('NONE', 'the obtained results are compared the ones obtained from XXXXX which is one of the most popular methods used in image segmentation , nondominated sorting XXXXX , and nondominated sorted particle swarm optimization', ['fuzzy c-means', 'genetic algorithm'])\n", "('method used for task', 'the obtained results are compared the ones obtained from XXXXX which is one of the most popular methods used in image segmentation , nondominated sorting genetic algorithm , and nondominated sorted XXXXX', ['fuzzy c-means', 'particle swarm optimization'])\n", "('NONE', 'the obtained results are compared the ones obtained from fuzzy c-means which is one of the most popular methods used in XXXXX , nondominated sorting XXXXX , and nondominated sorted particle swarm optimization', ['image segmentation', 'genetic algorithm'])\n", "('method used for task', 'the obtained results are compared the ones obtained from fuzzy c-means which is one of the most popular methods used in XXXXX , nondominated sorting genetic algorithm , and nondominated sorted XXXXX', ['image segmentation', 'particle swarm optimization'])\n", "('method used for task', 'the obtained results are compared the ones obtained from fuzzy c-means which is one of the most popular methods used in image segmentation , nondominated sorting XXXXX , and nondominated sorted XXXXX', ['genetic algorithm', 'particle swarm optimization'])\n", "('NONE', 'better performance than existing XXXXX and other approaches such as XXXXX and fcm clustering', ['k-means clustering', 'cosine similarity'])\n", "('NONE', 'pca proved effective in creating a XXXXX of XXXXX', ['digital signature', 'network traffic'])\n", "('method used for task', 'results pertaining to XXXXX and XXXXX are encouraging', ['false alarm', 'accuracy rate'])\n", "('NONE', 'flexibility to fix the XXXXX in h∞ loop shaping XXXXX', ['controller design', 'controller structure'])\n", "('method used for task', 'proposing a XXXXX based on gbmo for XXXXX', ['hybrid algorithm', 'image classification'])\n", "('method used for task', 'non-linearity between XXXXX ( rss ) and distance is modeled using fuzzy logic system ( fls ) to reduce the XXXXX and further optimized by hpso and bbo to minimize the error', ['received signal strength', 'computational complexity'])\n", "('NONE', 'due to high XXXXX of simulations , kriging XXXXX are built', ['computational cost', 'surrogate models'])\n", "('NONE', 'particle swarm algorithm with adaptive XXXXX optimizes XXXXX', ['constraint handling', 'surrogate model'])\n", "('NONE', 'XXXXX with adaptive XXXXX optimizes surrogate model', ['constraint handling', 'particle swarm algorithm'])\n", "('NONE', 'XXXXX with adaptive constraint handling optimizes XXXXX', ['surrogate model', 'particle swarm algorithm'])\n", "('method used for task', 'the XXXXX relies on human intuition and XXXXX', ['k-means clustering', 'industry standard'])\n", "('method used for task', 'a XXXXX ( ma-sat ) is proposed for XXXXX in networks', ['memetic algorithm', 'community detection'])\n", "('NONE', 'effective XXXXX in selecting the relevant XXXXX of cancer subtypes', ['clustering techniques', 'gene expression'])\n", "('NONE', 'it combines XXXXX with estimation of XXXXX', ['particle swarm optimization', 'distribution algorithm'])\n", "('NONE', 'a conditional spatial XXXXX ( csfcm ) XXXXX to improve the robustness of the conventional fcm algorithm is presented', ['fuzzy c-means', 'clustering algorithm'])\n", "('NONE', 'a conditional spatial XXXXX ( csfcm ) clustering algorithm to improve the robustness of the conventional XXXXX is presented', ['fuzzy c-means', 'fcm algorithm'])\n", "('NONE', 'a conditional spatial fuzzy c-means ( csfcm ) XXXXX to improve the robustness of the conventional XXXXX is presented', ['clustering algorithm', 'fcm algorithm'])\n", "('NONE', 'the method incorporates conditional affects and XXXXX into the XXXXX', ['spatial information', 'membership functions'])\n", "('NONE', 'this algorithm is based on the full XXXXX of XXXXX', ['bayesian approach', 'artificial neural networks'])\n", "('method used for task', 'XXXXX are integrated with gas and XXXXX', ['monte carlo methods', 'fuzzy membership functions'])\n", "('method used for task', 'proposed algorithm is applied to XXXXX and XXXXX in the context of bnns', ['time series', 'regression analysis'])\n", "('method used for task', 'XXXXX and XXXXX are roped as a level two classifier for good classification accuracy', ['nonlinear models', 'neural networks'])\n", "('method used for task', 'XXXXX and neural networks are roped as a level two classifier for good XXXXX', ['nonlinear models', 'classification accuracy'])\n", "('method used for task', 'nonlinear models and XXXXX are roped as a level two classifier for good XXXXX', ['neural networks', 'classification accuracy'])\n", "('method used for task', 'the actual XXXXX is carried out by XXXXX', ['time series prediction', 'neural networks'])\n", "('NONE', 'we integrate the XXXXX module into the neverstop XXXXX . the XXXXX architecture completes the integration and modeling of the traffic control systems', ['system design', 'fuzzy control'])\n", "('NONE', 'we integrate the XXXXX into the neverstop XXXXX . the fuzzy control architecture completes the integration and modeling of the traffic control systems', ['system design', 'fuzzy control module'])\n", "('NONE', 'we integrate the XXXXX module into the neverstop system design . the XXXXX architecture completes the integration and modeling of the traffic control systems', ['fuzzy control', 'fuzzy control module'])\n", "('NONE', 'we present a XXXXX illustrating how the average XXXXX is derived . the involvement amplifies the neverstop system and facilitates the fuzzy control module', ['genetic algorithm', 'waiting time'])\n", "('method used for task', 'we present a XXXXX illustrating how the average waiting time is derived . the involvement amplifies the neverstop system and facilitates the XXXXX', ['genetic algorithm', 'fuzzy control module'])\n", "('method used for task', 'we present a genetic algorithm illustrating how the average XXXXX is derived . the involvement amplifies the neverstop system and facilitates the XXXXX', ['waiting time', 'fuzzy control module'])\n", "('method used for task', 'neverstop utilizes fuzzy control method and XXXXX to adjust the XXXXX for the traffic lights , consequently the average XXXXX can be significantly reduced', ['genetic algorithm', 'waiting time'])\n", "('method used for task', 'neverstop utilizes fuzzy control method and XXXXX to adjust the XXXXX for the traffic lights , consequently the average XXXXX can be significantly reduced', ['genetic algorithm', 'waiting time'])\n", "('method used for task', 'neverstop utilizes fuzzy control method and XXXXX to adjust the waiting time for the XXXXX , consequently the average waiting time can be significantly reduced', ['genetic algorithm', 'traffic lights'])\n", "('method used for task', 'neverstop utilizes fuzzy control method and genetic algorithm to adjust the XXXXX for the XXXXX , consequently the average XXXXX can be significantly reduced', ['waiting time', 'traffic lights'])\n", "('method used for task', 'neverstop utilizes fuzzy control method and genetic algorithm to adjust the XXXXX for the XXXXX , consequently the average XXXXX can be significantly reduced', ['waiting time', 'traffic lights'])\n", "('NONE', 'the XXXXX is enhanced in the proposed system for fast retrieval of hidden data from XXXXX', ['hidden markov model', 'video files'])\n", "('method used for task', 'XXXXX and retrieval processes are performed using the conditional states and state transition dynamics between the XXXXX', ['data embedding', 'video frames'])\n", "('NONE', 'it enhances the retrieval XXXXX with minimized XXXXX', ['data rate', 'computation cost'])\n", "('method used for task', 'this paper proposes a new XXXXX mechanism for basic XXXXX', ['abc algorithm', 'solution update'])\n", "('method used for task', 'the XXXXX rule is based on XXXXX', ['normal distribution', 'solution update'])\n", "('method used for task', 'a novel XXXXX , coa-ga , is developed by integrating cuckoo XXXXX ( coa ) and ga to enhance XXXXX', ['optimization algorithm', 'classification performance'])\n", "('method used for task', 'a novel XXXXX , coa-ga , is developed by integrating cuckoo XXXXX ( coa ) and ga to enhance XXXXX', ['optimization algorithm', 'classification performance'])\n", "('method used for task', 'it is further confirmed that traditional clustering does not have any impact on XXXXX and XXXXX', ['gene selection', 'classification performance'])\n", "('NONE', 'adaptive XXXXX selects the most appropriate operators in an XXXXX', ['evolutionary algorithm', 'operator selection'])\n", "('NONE', 'hybridizes the XXXXX with XXXXX , where XXXXX is applied to control the randomness step inside the XXXXX', ['firefly algorithm', 'simulated annealing'])\n", "('NONE', 'hybridizes the XXXXX with XXXXX , where XXXXX is applied to control the randomness step inside the XXXXX', ['firefly algorithm', 'simulated annealing'])\n", "('NONE', 'hybridizes the XXXXX with XXXXX , where XXXXX is applied to control the randomness step inside the XXXXX', ['simulated annealing', 'firefly algorithm'])\n", "('NONE', 'hybridizes the XXXXX with XXXXX , where XXXXX is applied to control the randomness step inside the XXXXX', ['simulated annealing', 'firefly algorithm'])\n", "('method used for task', 'a XXXXX is embedded within the XXXXX to better explore the search space', ['lévy flight', 'firefly algorithm'])\n", "('method used for task', 'a XXXXX is embedded within the firefly algorithm to better explore the XXXXX', ['lévy flight', 'search space'])\n", "('method used for task', 'a lévy flight is embedded within the XXXXX to better explore the XXXXX', ['firefly algorithm', 'search space'])\n", "('method used for task', 'a combination of firefly , XXXXX and XXXXX is investigated to further improve the solution', ['lévy flight', 'simulated annealing'])\n", "('method used for task', 'the XXXXX is integrated in the conventional XXXXX', ['similarity measure', 'fuzzy c-means algorithm'])\n", "('method used for task', 'a compilation of XXXXX approaches for solving the XXXXX', ['soft computing', 'protein structure prediction problem'])\n", "('NONE', 'a summary of the basic concepts in the XXXXX of XXXXX', ['protein structure prediction', 'research field'])\n", "('NONE', 'in this study , an urban XXXXX dataset received from the database of uci XXXXX was used as the urban XXXXX data', ['machine learning', 'land cover'])\n", "('method used for task', 'this paper aims to improve the performance of XXXXX for solving XXXXX', ['harmony search algorithm', 'multi-objective optimization problems'])\n", "('method used for task', 'secondly , the proposed algorithm is applied to solve many classical XXXXX and it is also compared with other XXXXX', ['benchmark problems', 'multi-objective evolutionary algorithms'])\n", "('NONE', 'XXXXX experiments show that the controller proposed ( based on XXXXX ) is stable', ['fuzzy logic', 'lane changes'])\n", "('NONE', 'three XXXXX namely , iterative bi-section shooting ( ibss ) , XXXXX ( ann ) , and the novel new hybrid ann–ibss , were proposed', ['artificial neural network', 'parameter estimation methods'])\n", "('NONE', 'this paper has presented the our new XXXXX with all results in qos ( quality of XXXXX ) of wsn ( wireless sensor nodes ) and it is compared with leach , sep , genetic hcr and erp routing protocols', ['routing protocol', 'service metrics'])\n", "('NONE', 'three XXXXX are considered : XXXXX , power loss , voltage deviation and system loadability', ['objective functions', 'fuel cost'])\n", "('NONE', 'three XXXXX are considered : fuel cost , XXXXX , voltage deviation and system loadability', ['objective functions', 'power loss'])\n", "('NONE', 'three XXXXX are considered : fuel cost , power loss , XXXXX and system loadability', ['objective functions', 'voltage deviation'])\n", "('NONE', 'three objective functions are considered : XXXXX , XXXXX , voltage deviation and system loadability', ['fuel cost', 'power loss'])\n", "('NONE', 'three objective functions are considered : XXXXX , power loss , XXXXX and system loadability', ['fuel cost', 'voltage deviation'])\n", "('NONE', 'three objective functions are considered : fuel cost , XXXXX , XXXXX and system loadability', ['power loss', 'voltage deviation'])\n", "('NONE', 'modified XXXXX using XXXXX and external archive for multi-objective optimization', ['bat algorithm', 'inertia weight'])\n", "('method used for task', 'modified XXXXX using inertia weight and external archive for XXXXX', ['bat algorithm', 'multi-objective optimization'])\n", "('method used for task', 'modified bat algorithm using XXXXX and external archive for XXXXX', ['inertia weight', 'multi-objective optimization'])\n", "('method used for task', \"single-pass XXXXX for reducing operator 's annotation effort during on-line XXXXX\", ['active learning', 'visual inspection'])\n", "('method used for task', \"single-pass XXXXX for reducing operator 's XXXXX during on-line visual inspection\", ['active learning', 'annotation effort'])\n", "('NONE', \"single-pass active learning for reducing operator 's XXXXX during on-line XXXXX\", ['visual inspection', 'annotation effort'])\n", "('method used for task', 'intelligent hardware is based on the XXXXX and XXXXX ( nn )', ['wavelet decomposition', 'neural network'])\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "('NONE', 'integration of XXXXX with XXXXX for policy iteration', ['reinforcement learning', 'gaussian processes'])\n", "('method used for task', 'based on the adaptive neuro-fuzzy XXXXX ( anfis ) approach , a hazard modeling and survival prediction system is developed to assist clinicians in prognostic assessment of patients with esophageal cancer and prediction of individual XXXXX', ['inference system', 'patient survival'])\n", "('method used for task', 'this may provide valuable prognostic information in addition to ajcc staging and aid the clinicians’ XXXXX for XXXXX', ['decision-making process', 'risk stratification'])\n", "('method used for task', 'ann shows good XXXXX as compared to svr , rfr and XXXXX', ['prediction performance', 'mlr models'])\n", "('method used for task', 'this type of XXXXX can help to represent XXXXX having certain order', ['soft sets', 'linguistic terms'])\n", "('NONE', 'in certain XXXXX these XXXXX and operations on them can be very helpful', ['soft sets', 'decision making problems'])\n", "('NONE', 'incorporation of XXXXX in single- and three-area XXXXX', ['pid controller', 'power systems'])\n", "('NONE', 'investigation of idd XXXXX in five-area XXXXX', ['power system', 'controller performance'])\n", "('method used for task', 'a novel profit-based XXXXX for XXXXX with svm is presented', ['churn prediction', 'feature selection method'])\n", "('method used for task', 'determining critical XXXXX rules and forming a flexible XXXXX for XXXXX', ['rule base', 'phishing detection'])\n", "('method used for task', 'determining critical XXXXX rules and forming a flexible XXXXX for XXXXX', ['rule base', 'phishing detection'])\n", "('NONE', 'this study introduces a new methodology to practitioners and academics in which XXXXX used , in the XXXXX using the efqm excellence model for dealing with imprecision and applying ahp & or in the priority setting of improvement projects', ['fuzzy logic', 'performance evaluation'])\n", "('NONE', 'this study introduces a new methodology to practitioners and academics in which XXXXX used , in the performance evaluation using the efqm excellence model for dealing with imprecision and applying ahp & or in the priority setting of XXXXX', ['fuzzy logic', 'improvement projects'])\n", "('NONE', 'this study introduces a new methodology to practitioners and academics in which fuzzy logic used , in the XXXXX using the efqm excellence model for dealing with imprecision and applying ahp & or in the priority setting of XXXXX', ['performance evaluation', 'improvement projects'])\n", "('NONE', 'first one is the use of XXXXX for efqm excellence model not only make easier the procedure of XXXXX but also the total score for the organization performance obtained will be closer to the real score', ['fuzzy logic', 'performance assessment'])\n", "('method used for task', 'combining asynchronously XXXXX and XXXXX is the best program', ['inertia weight', 'constriction factor'])\n", "('method used for task', 'XXXXX for automatic XXXXX repudiate the need of a human expert', ['decision trees', 'rule learning'])\n", "('method used for task', 'automatic fuzzy partitioning of the XXXXX to over XXXXX', ['feature space', 'linguistic terms'])\n", "('NONE', 'high true positive rate and lower false positive rate for the proposed fuzzy rule base classification . high XXXXX achieved by the proposed system over XXXXX', ['classification accuracy', 'learning algorithms'])\n", "('method used for task', 'an ecg heart beat XXXXX is proposed based on modified XXXXX', ['abc algorithm', 'classification method'])\n", "('method used for task', 'total 38 XXXXX is calculated , then most distinctive XXXXX is used', ['feature set', 'feature subset'])\n", "('method used for task', 'XXXXX is achieved as 99.30 % on examined XXXXX from mitbih db', ['classification accuracy', 'ecg data'])\n", "('NONE', 'XXXXX within the interval type-2 XXXXX', ['multiple criteria decision analysis', 'fuzzy environment'])\n", "('method used for task', 'XXXXX are designed to test the XXXXX', ['computational experiments', 'solution approach'])\n", "('method used for task', 'XXXXX hp-hgs for solving XXXXX', ['hybrid strategy', 'inverse problems'])\n", "('NONE', 'global step solved by hierarchical XXXXX ( hgs ) coupled with self-adaptive hp-finite XXXXX ( hp-fem )', ['genetic search', 'element method'])\n", "('method used for task', 'rule pools for XXXXX are generated by a rule-based XXXXX', ['stock trading', 'evolutionary algorithm'])\n", "('method used for task', 'XXXXX for XXXXX are generated by a rule-based evolutionary algorithm', ['stock trading', 'rule pools'])\n", "('method used for task', 'XXXXX for stock trading are generated by a rule-based XXXXX', ['evolutionary algorithm', 'rule pools'])\n", "('NONE', 'XXXXX using mlp selects appropriate XXXXX for trading decision', ['ensemble learning', 'rule pools'])\n", "('NONE', 'properties of the fuzzy linguistic XXXXX on discrete XXXXX are analysed', ['model based', 'fuzzy numbers'])\n", "('NONE', 'some advantages of the XXXXX on discrete XXXXX are pointed out', ['model based', 'fuzzy numbers'])\n", "('NONE', 'aiming at the chaotic behavior of pmsg under certain parameters , a method based on adhdp is proposed to track the maximum wind power point , which can make the system out of chaos and track the maximum power point stably , and the XXXXX of the complex XXXXX can be solved effectively', ['optimal control problems', 'nonlinear system'])\n", "('method used for task', 'neutrosophic XXXXX is proposed based on XXXXX', ['cost function', 'neutrosophic set'])\n", "('NONE', 'solving the XXXXX in structure designing and XXXXX', ['parameter optimization', 'machining process'])\n", "('NONE', 'the proposed algorithm has better XXXXX than several popular XXXXX', ['computational efficiency', 'soft computing algorithms'])\n", "('method used for task', 'considering new features namely , XXXXX , opening and reopening modes , XXXXX and transportation modes', ['production facilities', 'greenhouse gas'])\n", "('method used for task', 'considering new features namely , XXXXX , opening and reopening modes , greenhouse gas and XXXXX', ['production facilities', 'transportation modes'])\n", "('method used for task', 'considering new features namely , production facilities , opening and reopening modes , XXXXX and XXXXX', ['greenhouse gas', 'transportation modes'])\n", "('NONE', 'to improve the exploration and XXXXX of bat algorithm some modifications based on XXXXX on bat algorithm are studied', ['chaotic map', 'exploitation capability'])\n", "('NONE', 'a new XXXXX inspired by XXXXX was proposed', ['tumor growth', 'heuristic optimization algorithm'])\n", "('method used for task', 'an empirical XXXXX is presented for solving different XXXXX', ['comparison study', 'test functions'])\n", "('method used for task', 'formally describes a form of modeling that unifies XXXXX and XXXXX', ['agent-based modeling', 'complex networks'])\n", "('NONE', 'we propose a method to transform the lineage expression into directed acyclic graphs ( dags ) equivalently starting from the lineage expressed as boolean formulas for spj queries over XXXXX . specifically , we discuss the corresponding XXXXX and properties to guarantee that the graphical model can support effective probabilistic inferences in lineage processing theoretically', ['uncertain data', 'probabilistic semantics'])\n", "('method used for task', 'we propose a method to transform the lineage expression into directed acyclic graphs ( dags ) equivalently starting from the lineage expressed as boolean formulas for spj queries over XXXXX . specifically , we discuss the corresponding probabilistic semantics and properties to guarantee that the XXXXX can support effective probabilistic inferences in lineage processing theoretically', ['uncertain data', 'graphical model'])\n", "('method used for task', 'we propose a method to transform the lineage expression into directed acyclic graphs ( dags ) equivalently starting from the lineage expressed as boolean formulas for spj queries over uncertain data . specifically , we discuss the corresponding XXXXX and properties to guarantee that the XXXXX can support effective probabilistic inferences in lineage processing theoretically', ['probabilistic semantics', 'graphical model'])\n", "('method used for task', 'we develop an XXXXX to solve interrelated XXXXX', ['evolutionary approach', 'optimisation problems'])\n", "('method used for task', 'an exact analytical XXXXX is introduced for interval type-2 XXXXX with closed form inference methods', ['inversion procedure', 'tsk fuzzy logic systems'])\n", "('NONE', 'the proposed procedure enables the use of more freely designed interval type-2 XXXXX in exact inverse model based XXXXX', ['engineering applications', 'tsk fuzzy logic systems'])\n", "('NONE', 'a new bionic algorithm is proposed based on the XXXXX of XXXXX branching , regrowing and tropisms', ['incentive mechanism', 'plant root'])\n", "('method used for task', 'dual XXXXX transform ( dtcwt ) is used as a new feature extractor from forward and reverse XXXXX', ['doppler ultrasound signals', 'tree complex wavelet'])\n", "('method used for task', 'an approach combining XXXXX and XXXXX is proposed', ['simulated annealing', 'mathematical programming'])\n", "('NONE', 'classifier diversity is achieved by diversifying the XXXXX and varying the initialization of the weights of the XXXXX', ['neural network', 'input features'])\n", "('method used for task', 'we examine changes in output of the proposed fusion model with increasing XXXXX , for all the XXXXX', ['test cases', 'base classifiers'])\n", "('NONE', 'lipschitz XXXXX and copula characteristic of t-norms and implications are discussed and the robustness of rule-based XXXXX is investigated', ['fuzzy reasoning', 'aggregation property'])\n", "('NONE', 'lipschitz aggregation property and XXXXX of t-norms and implications are discussed and the robustness of rule-based XXXXX is investigated', ['fuzzy reasoning', 'copula characteristic'])\n", "('method used for task', 'lipschitz XXXXX and XXXXX of t-norms and implications are discussed and the robustness of rule-based fuzzy reasoning is investigated', ['aggregation property', 'copula characteristic'])\n", "('NONE', 'according to lipschitz XXXXX and copula characteristic of t-norms and implications , suitable t-norm and implication can be chosen to satisfy the need of robustness of XXXXX', ['fuzzy reasoning', 'aggregation property'])\n", "('NONE', 'according to lipschitz aggregation property and XXXXX of t-norms and implications , suitable t-norm and implication can be chosen to satisfy the need of robustness of XXXXX', ['fuzzy reasoning', 'copula characteristic'])\n", "('method used for task', 'according to lipschitz XXXXX and XXXXX of t-norms and implications , suitable t-norm and implication can be chosen to satisfy the need of robustness of fuzzy reasoning', ['aggregation property', 'copula characteristic'])\n", "('NONE', 'approach for the optimization of modular neural networks with the XXXXX using XXXXX for pattern recognition', ['gravitational search algorithm', 'fuzzy logic'])\n", "('NONE', 'approach for the optimization of modular neural networks with the XXXXX using fuzzy logic for XXXXX', ['gravitational search algorithm', 'pattern recognition'])\n", "('method used for task', 'approach for the optimization of modular neural networks with the gravitational search algorithm using XXXXX for XXXXX', ['fuzzy logic', 'pattern recognition'])\n", "('NONE', 'bespectacled XXXXX will reduce the XXXXX', ['eye image', 'iris segmentation accuracy'])\n", "('method used for task', 'XXXXX is performed as the XXXXX for frame detection', ['fuzzy logic', 'decision making'])\n", "('method used for task', 'eigenstructure based XXXXX for XXXXX', ['fault detection', 'nonlinear systems'])\n", "('method used for task', 'compared four XXXXX for learning XXXXX', ['fuzzy cognitive maps', 'stochastic optimization methods'])\n", "('method used for task', 'a novel XXXXX based neighborhood construction and a fast XXXXX', ['tree search', 'evaluation method'])\n", "('NONE', 'a novel XXXXX based XXXXX and a fast evaluation method', ['tree search', 'neighborhood construction'])\n", "('method used for task', 'a novel tree search based XXXXX and a fast XXXXX', ['evaluation method', 'neighborhood construction'])\n", "('NONE', 'the generating units’ contingencies and XXXXX are explicitly considered in the XXXXX of htss', ['price uncertainty', 'stochastic programming'])\n", "('method used for task', '18 XXXXX and two real-world problems are used in XXXXX', ['benchmark functions', 'experimental study'])\n", "('method used for task', '18 XXXXX and two XXXXX are used in experimental study', ['benchmark functions', 'real-world problems'])\n", "('NONE', '18 benchmark functions and two XXXXX are used in XXXXX', ['experimental study', 'real-world problems'])\n", "('NONE', 'presenting XXXXX in the sense of the proposed XXXXX', ['lyapunov function', 'adaptation laws'])\n", "('NONE', 'the XXXXX and the robustness of wsa were proven through a set of XXXXX', ['computational study', 'convergence capability'])\n", "('method used for task', 'an optimal XXXXX as used in dpc is prohibited due to the extremely large size of the XXXXX in this optimization problem', ['exhaustive search', 'search space'])\n", "('NONE', 'an optimal XXXXX as used in dpc is prohibited due to the extremely large size of the search space in this XXXXX', ['exhaustive search', 'optimization problem'])\n", "('NONE', 'an optimal exhaustive search as used in dpc is prohibited due to the extremely large size of the XXXXX in this XXXXX', ['search space', 'optimization problem'])\n", "('NONE', 'XXXXX ( XXXXX ) can be used as an alternative for this optimization problem to reduce the complexity of the search ( scheduling problem ) . the performance of the XXXXX with elitism and adaptive mutation is demonstrated to be near optimal as obtained with an exhaustive search . it has been demonstrated in this paper that ga achieves about 98–99 % of system sum rate as obtained with dpc with significant reduction in time and computational complexity', ['evolutionary algorithm', 'genetic algorithm'])\n", "('NONE', 'XXXXX ( genetic algorithm ) can be used as an alternative for this XXXXX to reduce the complexity of the search ( scheduling problem ) . the performance of the genetic algorithm with elitism and adaptive mutation is demonstrated to be near optimal as obtained with an exhaustive search . it has been demonstrated in this paper that ga achieves about 98–99 % of system sum rate as obtained with dpc with significant reduction in time and computational complexity', ['evolutionary algorithm', 'optimization problem'])\n", "('method used for task', 'XXXXX ( genetic algorithm ) can be used as an alternative for this optimization problem to reduce the complexity of the search ( XXXXX ) . the performance of the genetic algorithm with elitism and adaptive mutation is demonstrated to be near optimal as obtained with an exhaustive search . it has been demonstrated in this paper that ga achieves about 98–99 % of system sum rate as obtained with dpc with significant reduction in time and computational complexity', ['evolutionary algorithm', 'scheduling problem'])\n", "('NONE', 'XXXXX ( XXXXX ) can be used as an alternative for this optimization problem to reduce the complexity of the search ( scheduling problem ) . the performance of the XXXXX with elitism and adaptive mutation is demonstrated to be near optimal as obtained with an exhaustive search . it has been demonstrated in this paper that ga achieves about 98–99 % of system sum rate as obtained with dpc with significant reduction in time and computational complexity', ['evolutionary algorithm', 'genetic algorithm'])\n", "('method used for task', 'XXXXX ( genetic algorithm ) can be used as an alternative for this optimization problem to reduce the complexity of the search ( scheduling problem ) . the performance of the genetic algorithm with elitism and adaptive mutation is demonstrated to be near optimal as obtained with an XXXXX . it has been demonstrated in this paper that ga achieves about 98–99 % of system sum rate as obtained with dpc with significant reduction in time and computational complexity', ['evolutionary algorithm', 'exhaustive search'])\n", "('NONE', 'XXXXX ( genetic algorithm ) can be used as an alternative for this optimization problem to reduce the complexity of the search ( scheduling problem ) . the performance of the genetic algorithm with elitism and adaptive mutation is demonstrated to be near optimal as obtained with an exhaustive search . it has been demonstrated in this paper that ga achieves about 98–99 % of system sum rate as obtained with dpc with significant reduction in time and XXXXX', ['evolutionary algorithm', 'computational complexity'])\n", "('NONE', 'evolutionary algorithm ( XXXXX ) can be used as an alternative for this XXXXX to reduce the complexity of the search ( scheduling problem ) . the performance of the XXXXX with elitism and adaptive mutation is demonstrated to be near optimal as obtained with an exhaustive search . it has been demonstrated in this paper that ga achieves about 98–99 % of system sum rate as obtained with dpc with significant reduction in time and computational complexity', ['genetic algorithm', 'optimization problem'])\n", "('method used for task', 'evolutionary algorithm ( XXXXX ) can be used as an alternative for this optimization problem to reduce the complexity of the search ( XXXXX ) . the performance of the XXXXX with elitism and adaptive mutation is demonstrated to be near optimal as obtained with an exhaustive search . it has been demonstrated in this paper that ga achieves about 98–99 % of system sum rate as obtained with dpc with significant reduction in time and computational complexity', ['genetic algorithm', 'scheduling problem'])\n", "('NONE', 'evolutionary algorithm ( XXXXX ) can be used as an alternative for this optimization problem to reduce the complexity of the search ( scheduling problem ) . the performance of the XXXXX with elitism and adaptive mutation is demonstrated to be near optimal as obtained with an XXXXX . it has been demonstrated in this paper that ga achieves about 98–99 % of system sum rate as obtained with dpc with significant reduction in time and computational complexity', ['genetic algorithm', 'exhaustive search'])\n", "('NONE', 'evolutionary algorithm ( XXXXX ) can be used as an alternative for this optimization problem to reduce the complexity of the search ( scheduling problem ) . the performance of the XXXXX with elitism and adaptive mutation is demonstrated to be near optimal as obtained with an exhaustive search . it has been demonstrated in this paper that ga achieves about 98–99 % of system sum rate as obtained with dpc with significant reduction in time and XXXXX', ['genetic algorithm', 'computational complexity'])\n", "('NONE', 'evolutionary algorithm ( genetic algorithm ) can be used as an alternative for this XXXXX to reduce the complexity of the search ( XXXXX ) . the performance of the genetic algorithm with elitism and adaptive mutation is demonstrated to be near optimal as obtained with an exhaustive search . it has been demonstrated in this paper that ga achieves about 98–99 % of system sum rate as obtained with dpc with significant reduction in time and computational complexity', ['optimization problem', 'scheduling problem'])\n", "('NONE', 'evolutionary algorithm ( XXXXX ) can be used as an alternative for this XXXXX to reduce the complexity of the search ( scheduling problem ) . the performance of the XXXXX with elitism and adaptive mutation is demonstrated to be near optimal as obtained with an exhaustive search . it has been demonstrated in this paper that ga achieves about 98–99 % of system sum rate as obtained with dpc with significant reduction in time and computational complexity', ['optimization problem', 'genetic algorithm'])\n", "('method used for task', 'evolutionary algorithm ( genetic algorithm ) can be used as an alternative for this XXXXX to reduce the complexity of the search ( scheduling problem ) . the performance of the genetic algorithm with elitism and adaptive mutation is demonstrated to be near optimal as obtained with an XXXXX . it has been demonstrated in this paper that ga achieves about 98–99 % of system sum rate as obtained with dpc with significant reduction in time and computational complexity', ['optimization problem', 'exhaustive search'])\n", "('NONE', 'evolutionary algorithm ( genetic algorithm ) can be used as an alternative for this XXXXX to reduce the complexity of the search ( scheduling problem ) . the performance of the genetic algorithm with elitism and adaptive mutation is demonstrated to be near optimal as obtained with an exhaustive search . it has been demonstrated in this paper that ga achieves about 98–99 % of system sum rate as obtained with dpc with significant reduction in time and XXXXX', ['optimization problem', 'computational complexity'])\n", "('method used for task', 'evolutionary algorithm ( XXXXX ) can be used as an alternative for this optimization problem to reduce the complexity of the search ( XXXXX ) . the performance of the XXXXX with elitism and adaptive mutation is demonstrated to be near optimal as obtained with an exhaustive search . it has been demonstrated in this paper that ga achieves about 98–99 % of system sum rate as obtained with dpc with significant reduction in time and computational complexity', ['scheduling problem', 'genetic algorithm'])\n", "('method used for task', 'evolutionary algorithm ( genetic algorithm ) can be used as an alternative for this optimization problem to reduce the complexity of the search ( XXXXX ) . the performance of the genetic algorithm with elitism and adaptive mutation is demonstrated to be near optimal as obtained with an XXXXX . it has been demonstrated in this paper that ga achieves about 98–99 % of system sum rate as obtained with dpc with significant reduction in time and computational complexity', ['scheduling problem', 'exhaustive search'])\n", "('NONE', 'evolutionary algorithm ( genetic algorithm ) can be used as an alternative for this optimization problem to reduce the complexity of the search ( XXXXX ) . the performance of the genetic algorithm with elitism and adaptive mutation is demonstrated to be near optimal as obtained with an exhaustive search . it has been demonstrated in this paper that ga achieves about 98–99 % of system sum rate as obtained with dpc with significant reduction in time and XXXXX', ['scheduling problem', 'computational complexity'])\n", "('NONE', 'evolutionary algorithm ( XXXXX ) can be used as an alternative for this optimization problem to reduce the complexity of the search ( scheduling problem ) . the performance of the XXXXX with elitism and adaptive mutation is demonstrated to be near optimal as obtained with an XXXXX . it has been demonstrated in this paper that ga achieves about 98–99 % of system sum rate as obtained with dpc with significant reduction in time and computational complexity', ['genetic algorithm', 'exhaustive search'])\n", "('NONE', 'evolutionary algorithm ( XXXXX ) can be used as an alternative for this optimization problem to reduce the complexity of the search ( scheduling problem ) . the performance of the XXXXX with elitism and adaptive mutation is demonstrated to be near optimal as obtained with an exhaustive search . it has been demonstrated in this paper that ga achieves about 98–99 % of system sum rate as obtained with dpc with significant reduction in time and XXXXX', ['genetic algorithm', 'computational complexity'])\n", "('NONE', 'evolutionary algorithm ( genetic algorithm ) can be used as an alternative for this optimization problem to reduce the complexity of the search ( scheduling problem ) . the performance of the genetic algorithm with elitism and adaptive mutation is demonstrated to be near optimal as obtained with an XXXXX . it has been demonstrated in this paper that ga achieves about 98–99 % of system sum rate as obtained with dpc with significant reduction in time and XXXXX', ['exhaustive search', 'computational complexity'])\n", "('method used for task', 'novel XXXXX based on multi-gene XXXXX called gpfis-class', ['genetic fuzzy system', 'genetic programming'])\n", "('NONE', 'gpfis-class builds XXXXX premises by using multi-gene XXXXX', ['fuzzy rule', 'genetic programming'])\n", "('NONE', 'this paper address the bodyguard XXXXX ( bap ) through an XXXXX', ['allocation problem', 'evolutionary approach'])\n", "('NONE', 'we develop a XXXXX ( sa ) with special XXXXX and restart strategy for mc-top-mtw', ['simulated annealing', 'solution encoding'])\n", "('NONE', 'we develop a XXXXX ( sa ) with special solution encoding and XXXXX for mc-top-mtw', ['simulated annealing', 'restart strategy'])\n", "('method used for task', 'we develop a simulated annealing ( sa ) with special XXXXX and XXXXX for mc-top-mtw', ['solution encoding', 'restart strategy'])\n", "('NONE', 'experimental results demonstrate that mmtd has significantly performed better than state-of-the-art moeas such as moea/d , nsga-ii , amalgam , mopso : a proposal for multiple objective XXXXX , mosade , decmosa-sqp in terms of reducing the inverted generational distance ( igd ) metric and XXXXX on both zdt test problems [ 68 ] and cec’09 test instances', ['particle swarm optimization', 'time cost'])\n", "('NONE', 'to ensure good XXXXX , a XXXXX for manets must change its routing policy online to account for changes in network conditions and to deal with routing information imprecision', ['network performance', 'routing protocol'])\n", "('method used for task', 'the main focus of this paper is on the use of XXXXX and XXXXX', ['fuzzy logic', 'reinforcement learning'])\n", "('NONE', 'a fuzzy extension of a XXXXX based XXXXX for manets is presented', ['reinforcement learning', 'routing protocol'])\n", "('method used for task', 'dynamic XXXXX is more appropriate than XXXXX for adaptive energy aware routing in manets', ['fuzzy logic', 'reinforcement learning'])\n", "('method used for task', 'proposed XXXXX and variable reduction strategy reduces XXXXX', ['equality constraints', 'equality constraint'])\n", "('NONE', 'proposed equality constraint and XXXXX reduces XXXXX', ['equality constraints', 'variable reduction strategy'])\n", "('method used for task', 'proposed XXXXX and XXXXX reduces equality constraints', ['equality constraint', 'variable reduction strategy'])\n", "('method used for task', 'proposed XXXXX and XXXXX also reduces the variables of cops', ['equality constraint', 'variable reduction strategy'])\n", "('method used for task', 'a new improved XXXXX , one rank cuckoo search algorithm ( orcsa ) , is applied for solving economic XXXXX problem', ['meta-heuristic algorithm', 'load dispatch'])\n", "('method used for task', 'the advantages of the proposed orcsa are few XXXXX , high solution quality and fast XXXXX', ['control parameters', 'computational time'])\n", "('NONE', 'the advantages of the proposed orcsa are few XXXXX , high XXXXX and fast computational time', ['control parameters', 'solution quality'])\n", "('method used for task', 'the advantages of the proposed orcsa are few control parameters , high XXXXX and fast XXXXX', ['computational time', 'solution quality'])\n", "('method used for task', 'automatic XXXXX for XXXXX malignancy grading', ['clinical decision support system', 'breast cancer'])\n", "('method used for task', 'XXXXX is the most efficient XXXXX for this problem', ['particle swarm', 'search method'])\n", "('method used for task', 'we introduce three kinds of XXXXX for dual XXXXX , which the corresponding support measures can be obtained', ['distance measures', 'hesitant fuzzy sets'])\n", "('method used for task', 'we propose several power aggregation operators on dual XXXXX , study their properties and give some specific dual hesitant XXXXX', ['hesitant fuzzy sets', 'fuzzy aggregation operators'])\n", "('method used for task', 'the proposed multiple inputs and outputs ( mio ) classification method designated as the fvm-index method integrates XXXXX ( fst ) , variable precision rough set ( vprs ) theory , and a modified XXXXX ( mcvi ) function , and is designed specifically to filter out the uncertainty and inaccuracy inherent in the surveyed mio real-valued dataset ; thereby improving the classification performance', ['fuzzy set theory', 'cluster validity index'])\n", "('method used for task', 'the proposed multiple inputs and outputs ( mio ) classification method designated as the fvm-index method integrates XXXXX ( fst ) , variable precision rough set ( vprs ) theory , and a modified cluster validity index ( mcvi ) function , and is designed specifically to filter out the uncertainty and inaccuracy inherent in the surveyed mio real-valued dataset ; thereby improving the XXXXX', ['fuzzy set theory', 'classification performance'])\n", "('NONE', 'the proposed multiple inputs and outputs ( mio ) XXXXX designated as the fvm-index method integrates XXXXX ( fst ) , variable precision rough set ( vprs ) theory , and a modified cluster validity index ( mcvi ) function , and is designed specifically to filter out the uncertainty and inaccuracy inherent in the surveyed mio real-valued dataset ; thereby improving the classification performance', ['fuzzy set theory', 'classification method'])\n", "('method used for task', 'the proposed multiple inputs and outputs ( mio ) classification method designated as the fvm-index method integrates fuzzy set theory ( fst ) , variable precision rough set ( vprs ) theory , and a modified XXXXX ( mcvi ) function , and is designed specifically to filter out the uncertainty and inaccuracy inherent in the surveyed mio real-valued dataset ; thereby improving the XXXXX', ['cluster validity index', 'classification performance'])\n", "('NONE', 'the proposed multiple inputs and outputs ( mio ) XXXXX designated as the fvm-index method integrates fuzzy set theory ( fst ) , variable precision rough set ( vprs ) theory , and a modified XXXXX ( mcvi ) function , and is designed specifically to filter out the uncertainty and inaccuracy inherent in the surveyed mio real-valued dataset ; thereby improving the classification performance', ['cluster validity index', 'classification method'])\n", "('method used for task', 'the proposed multiple inputs and outputs ( mio ) XXXXX designated as the fvm-index method integrates fuzzy set theory ( fst ) , variable precision rough set ( vprs ) theory , and a modified cluster validity index ( mcvi ) function , and is designed specifically to filter out the uncertainty and inaccuracy inherent in the surveyed mio real-valued dataset ; thereby improving the XXXXX', ['classification performance', 'classification method'])\n", "('NONE', 'XXXXX leverages on XXXXX ( rl ) to enhance network security', ['cognitive radio', 'reinforcement learning'])\n", "('NONE', 'we XXXXX by considering the asset returns as XXXXX', ['model uncertainty', 'trapezoidal fuzzy numbers'])\n", "('NONE', 'we XXXXX by considering the XXXXX as trapezoidal fuzzy numbers', ['model uncertainty', 'asset returns'])\n", "('NONE', 'we model uncertainty by considering the XXXXX as XXXXX', ['trapezoidal fuzzy numbers', 'asset returns'])\n", "('NONE', \"the model is validated by applying it to a real XXXXX in ecuador 's XXXXX\", ['case study', 'ict sector'])\n", "('method used for task', 'the XXXXX is solved using a XXXXX', ['optimization problem', 'genetic algorithm'])\n", "('method used for task', 'achieved balanced clustering for enhanced XXXXX and minimum XXXXX', ['network lifetime', 'packet loss'])\n", "('method used for task', 'the XXXXX function is implemented as a XXXXX', ['neural network', 'pairwise preference'])\n", "('method used for task', 'fuzzy least squares regression ( flsr ) and XXXXX based modeling methods , switching fuzzy c-regression ( sfcr ) and takagi–sugeno ( ts ) XXXXX , were used for modeling of multi-response experiment data with replicated response measures', ['fuzzy clustering', 'fuzzy model'])\n", "('method used for task', 'fuzzy least squares regression ( flsr ) and XXXXX based modeling methods , switching fuzzy c-regression ( sfcr ) and takagi–sugeno ( ts ) fuzzy model , were used for modeling of multi-response XXXXX with replicated response measures', ['fuzzy clustering', 'experiment data'])\n", "('NONE', 'fuzzy least squares regression ( flsr ) and XXXXX based modeling methods , switching fuzzy c-regression ( sfcr ) and takagi–sugeno ( ts ) fuzzy model , were used for modeling of multi-response experiment data with replicated XXXXX', ['fuzzy clustering', 'response measures'])\n", "('NONE', 'fuzzy least squares regression ( flsr ) and fuzzy clustering based modeling methods , switching fuzzy c-regression ( sfcr ) and takagi–sugeno ( ts ) XXXXX , were used for modeling of multi-response XXXXX with replicated response measures', ['fuzzy model', 'experiment data'])\n", "('NONE', 'fuzzy least squares regression ( flsr ) and fuzzy clustering based modeling methods , switching fuzzy c-regression ( sfcr ) and takagi–sugeno ( ts ) XXXXX , were used for modeling of multi-response experiment data with replicated XXXXX', ['fuzzy model', 'response measures'])\n", "('NONE', 'fuzzy least squares regression ( flsr ) and fuzzy clustering based modeling methods , switching fuzzy c-regression ( sfcr ) and takagi–sugeno ( ts ) fuzzy model , were used for modeling of multi-response XXXXX with replicated XXXXX', ['experiment data', 'response measures'])\n", "('method used for task', 'it was seen that the sfcr had the better XXXXX rather than flsr and ts fuzzy model according to the XXXXX square error ( rmse )', ['prediction performance', 'root mean'])\n", "('method used for task', 'we introduce a suitable distance formula for XXXXX , which is compared with the XXXXX , euclidean distance , XXXXX based on hausdorff metric and signed distance', ['interval type-2 fuzzy sets', 'hamming distance'])\n", "('method used for task', 'we introduce a suitable distance formula for XXXXX , which is compared with the hamming distance , XXXXX , hamming distance based on hausdorff metric and signed distance', ['interval type-2 fuzzy sets', 'euclidean distance'])\n", "('method used for task', 'we introduce a suitable distance formula for XXXXX , which is compared with the XXXXX , euclidean distance , XXXXX based on hausdorff metric and signed distance', ['interval type-2 fuzzy sets', 'hamming distance'])\n", "('method used for task', 'we introduce a suitable XXXXX for XXXXX , which is compared with the hamming distance , euclidean distance , hamming distance based on hausdorff metric and signed distance', ['interval type-2 fuzzy sets', 'distance formula'])\n", "('NONE', 'we introduce a suitable distance formula for interval type-2 fuzzy sets , which is compared with the XXXXX , XXXXX , XXXXX based on hausdorff metric and signed distance', ['hamming distance', 'euclidean distance'])\n", "('method used for task', 'we introduce a suitable XXXXX for interval type-2 fuzzy sets , which is compared with the XXXXX , euclidean distance , XXXXX based on hausdorff metric and signed distance', ['hamming distance', 'distance formula'])\n", "('NONE', 'we introduce a suitable distance formula for interval type-2 fuzzy sets , which is compared with the XXXXX , XXXXX , XXXXX based on hausdorff metric and signed distance', ['euclidean distance', 'hamming distance'])\n", "('method used for task', 'we introduce a suitable XXXXX for interval type-2 fuzzy sets , which is compared with the hamming distance , XXXXX , hamming distance based on hausdorff metric and signed distance', ['euclidean distance', 'distance formula'])\n", "('method used for task', 'we introduce a suitable XXXXX for interval type-2 fuzzy sets , which is compared with the XXXXX , euclidean distance , XXXXX based on hausdorff metric and signed distance', ['hamming distance', 'distance formula'])\n", "('method used for task', 'based on the proposed XXXXX , we propose a hierarchical clustering-based method to solve a XXXXX and find the proximity of the suppliers', ['distance formula', 'supplier selection problem'])\n", "('method used for task', 'to illustrate the applicability of the proposed method , two examples are illustrated . the first example is a XXXXX in supplier selection problem . the next one is an example taken from the literature . then , to test the hierarchical clustering-based method and compare the obtained results by two other methods , a XXXXX using experimental analysis was designed', ['case study', 'comparative study'])\n", "('method used for task', 'to illustrate the applicability of the proposed method , two examples are illustrated . the first example is a XXXXX in supplier selection problem . the next one is an example taken from the literature . then , to test the hierarchical clustering-based method and compare the obtained results by two other methods , a comparative study using XXXXX was designed', ['case study', 'experimental analysis'])\n", "('NONE', 'to illustrate the applicability of the proposed method , two examples are illustrated . the first example is a XXXXX in XXXXX . the next one is an example taken from the literature . then , to test the hierarchical clustering-based method and compare the obtained results by two other methods , a comparative study using experimental analysis was designed', ['case study', 'supplier selection problem'])\n", "('NONE', 'to illustrate the applicability of the proposed method , two examples are illustrated . the first example is a case study in supplier selection problem . the next one is an example taken from the literature . then , to test the hierarchical clustering-based method and compare the obtained results by two other methods , a XXXXX using XXXXX was designed', ['comparative study', 'experimental analysis'])\n", "('method used for task', 'to illustrate the applicability of the proposed method , two examples are illustrated . the first example is a case study in XXXXX . the next one is an example taken from the literature . then , to test the hierarchical clustering-based method and compare the obtained results by two other methods , a XXXXX using experimental analysis was designed', ['comparative study', 'supplier selection problem'])\n", "('method used for task', 'to illustrate the applicability of the proposed method , two examples are illustrated . the first example is a case study in XXXXX . the next one is an example taken from the literature . then , to test the hierarchical clustering-based method and compare the obtained results by two other methods , a comparative study using XXXXX was designed', ['experimental analysis', 'supplier selection problem'])\n", "('method used for task', 'the novelties of our work are in both theory and application . we propose a new distance formula for XXXXX . based on the proposed distance formula , we propose a new XXXXX to solve a supplier selection problem and finally to illustrate the applicability of the proposed method , a case study is used', ['interval type-2 fuzzy sets', 'clustering method'])\n", "('method used for task', 'the novelties of our work are in both theory and application . we propose a new distance formula for XXXXX . based on the proposed distance formula , we propose a new clustering method to solve a supplier selection problem and finally to illustrate the applicability of the proposed method , a XXXXX is used', ['interval type-2 fuzzy sets', 'case study'])\n", "('method used for task', 'the novelties of our work are in both theory and application . we propose a new XXXXX for XXXXX . based on the proposed XXXXX , we propose a new clustering method to solve a supplier selection problem and finally to illustrate the applicability of the proposed method , a case study is used', ['interval type-2 fuzzy sets', 'distance formula'])\n", "('method used for task', 'the novelties of our work are in both theory and application . we propose a new XXXXX for XXXXX . based on the proposed XXXXX , we propose a new clustering method to solve a supplier selection problem and finally to illustrate the applicability of the proposed method , a case study is used', ['interval type-2 fuzzy sets', 'distance formula'])\n", "('method used for task', 'the novelties of our work are in both theory and application . we propose a new distance formula for XXXXX . based on the proposed distance formula , we propose a new clustering method to solve a XXXXX and finally to illustrate the applicability of the proposed method , a case study is used', ['interval type-2 fuzzy sets', 'supplier selection problem'])\n", "('method used for task', 'the novelties of our work are in both theory and application . we propose a new distance formula for interval type-2 fuzzy sets . based on the proposed distance formula , we propose a new XXXXX to solve a supplier selection problem and finally to illustrate the applicability of the proposed method , a XXXXX is used', ['clustering method', 'case study'])\n", "('method used for task', 'the novelties of our work are in both theory and application . we propose a new XXXXX for interval type-2 fuzzy sets . based on the proposed XXXXX , we propose a new XXXXX to solve a supplier selection problem and finally to illustrate the applicability of the proposed method , a case study is used', ['clustering method', 'distance formula'])\n", "('method used for task', 'the novelties of our work are in both theory and application . we propose a new XXXXX for interval type-2 fuzzy sets . based on the proposed XXXXX , we propose a new XXXXX to solve a supplier selection problem and finally to illustrate the applicability of the proposed method , a case study is used', ['clustering method', 'distance formula'])\n", "('method used for task', 'the novelties of our work are in both theory and application . we propose a new distance formula for interval type-2 fuzzy sets . based on the proposed distance formula , we propose a new XXXXX to solve a XXXXX and finally to illustrate the applicability of the proposed method , a case study is used', ['clustering method', 'supplier selection problem'])\n", "('method used for task', 'the novelties of our work are in both theory and application . we propose a new XXXXX for interval type-2 fuzzy sets . based on the proposed XXXXX , we propose a new clustering method to solve a supplier selection problem and finally to illustrate the applicability of the proposed method , a XXXXX is used', ['case study', 'distance formula'])\n", "('method used for task', 'the novelties of our work are in both theory and application . we propose a new XXXXX for interval type-2 fuzzy sets . based on the proposed XXXXX , we propose a new clustering method to solve a supplier selection problem and finally to illustrate the applicability of the proposed method , a XXXXX is used', ['case study', 'distance formula'])\n", "('NONE', 'the novelties of our work are in both theory and application . we propose a new distance formula for interval type-2 fuzzy sets . based on the proposed distance formula , we propose a new clustering method to solve a XXXXX and finally to illustrate the applicability of the proposed method , a XXXXX is used', ['case study', 'supplier selection problem'])\n", "('method used for task', 'the novelties of our work are in both theory and application . we propose a new XXXXX for interval type-2 fuzzy sets . based on the proposed XXXXX , we propose a new clustering method to solve a XXXXX and finally to illustrate the applicability of the proposed method , a case study is used', ['distance formula', 'supplier selection problem'])\n", "('method used for task', 'the novelties of our work are in both theory and application . we propose a new XXXXX for interval type-2 fuzzy sets . based on the proposed XXXXX , we propose a new clustering method to solve a XXXXX and finally to illustrate the applicability of the proposed method , a case study is used', ['distance formula', 'supplier selection problem'])\n", "('method used for task', 'large scale problems require large XXXXX and XXXXX', ['population size', 'computational cost'])\n", "('NONE', 'XXXXX require large XXXXX and computational cost', ['population size', 'large scale problems'])\n", "('method used for task', 'XXXXX require large population size and XXXXX', ['computational cost', 'large scale problems'])\n", "('NONE', 'new justification of the XXXXX by XXXXX', ['theoretical model', 'hesitant fuzzy sets'])\n", "('NONE', 'it decreases the uncertainty and the XXXXX in XXXXX', ['information loss', 'group decision making'])\n", "('method used for task', 'further study can focus on interval type 2 XXXXX to develop a XXXXX', ['fuzzy set', 'decision support system'])\n", "('NONE', 'the interpretability measure is based on the XXXXX of three components : the average length of rules , the number of active XXXXX , and the number of active inputs of the system', ['arithmetic mean', 'fuzzy sets'])\n", "('NONE', 'a novel method for XXXXX with intuitionistic XXXXX is proposed and applied to radio frequency identification ( rfid ) technology selection', ['group decision making', 'fuzzy preference relations'])\n", "('method used for task', 'a novel method for XXXXX with intuitionistic fuzzy preference relations is proposed and applied to radio frequency identification ( rfid ) XXXXX', ['group decision making', 'technology selection'])\n", "('method used for task', 'a novel method for group decision making with intuitionistic XXXXX is proposed and applied to radio frequency identification ( rfid ) XXXXX', ['fuzzy preference relations', 'technology selection'])\n", "('NONE', 'we introduce a XXXXX considering articles on XXXXX in business published in last two decades', ['literature review', 'artificial neural networks'])\n", "('NONE', 'a synthetic XXXXX using XXXXX is presented', ['evaluation method', 'triangular fuzzy number'])\n", "('NONE', 'we get a speech signal model on a XXXXX by XXXXX', ['reconstructed phase space', 'chaotic time series'])\n", "('method used for task', 'presents an XXXXX for finding XXXXX in different dimensions for each particle', ['adaptive method', 'inertia weight'])\n", "('method used for task', 'XXXXX is employed to deal with fuzziness , and a set of XXXXX is obtained', ['ranking function', 'efficient solutions'])\n", "('NONE', 'proposed a system for grammar inference by utilizing the mask-fill XXXXX and boolean based procedure with minimum XXXXX principle', ['reproduction operators', 'description length'])\n", "('NONE', 'we design a new XXXXX which takes into account two term , the XXXXX and the class separability distance', ['objective function', 'classification error rate'])\n", "('method used for task', 'the framework also considers XXXXX and XXXXX in dea models', ['high-dimensional data', 'missing values'])\n", "('NONE', 'the framework also considers XXXXX and missing values in XXXXX', ['high-dimensional data', 'dea models'])\n", "('NONE', 'the framework also considers high-dimensional data and XXXXX in XXXXX', ['missing values', 'dea models'])\n", "('NONE', 'a dimension-reduction method improves the XXXXX of the XXXXX', ['discrimination power', 'dea model'])\n", "('NONE', 'a preference ratio method ranks the interval XXXXX in the XXXXX', ['fuzzy environment', 'efficiency scores'])\n", "('NONE', 'equipping the svr model with ann as XXXXX has improved the XXXXX', ['kernel function', 'model accuracy'])\n", "('method used for task', 'a maximizing XXXXX is established for determining XXXXX', ['optimization model', 'criteria weights'])\n", "('method used for task', 'an experimental hardware , including the proposed XXXXX method is set up . the experimental results have confirmed that the XXXXX method exhibits satisfactory predicting performance for svi', ['soft computing', 'soft computing method'])\n", "('method used for task', 'an experimental hardware , including the proposed XXXXX method is set up . the experimental results have confirmed that the XXXXX method exhibits satisfactory predicting performance for svi', ['soft computing', 'soft computing method'])\n", "('method used for task', 'XXXXX is the graphical analysis by which the behavior between certain XXXXX of the antenna can be visually and numerically analyzed', ['curve fitting', 'design parameters'])\n", "('method used for task', 'XXXXX is the individual element called the particles form a swarm , all particles attains one initial position randomly at the start and then they fly through the genotypic space to find the XXXXX of the problem', ['particle swarm optimization', 'optimal solution'])\n", "('method used for task', 'we present a study on the adaptation of the island sizes of a distributed XXXXX to the XXXXX of the nodes of an heterogeneous cluster', ['evolutionary algorithm', 'computational power'])\n", "('NONE', 'firstly , the XXXXX uses XXXXX ( ga ) with a high exploration power', ['hybrid method', 'genetic algorithm'])\n", "('NONE', 'detail flow diagram of processes for obtaining the proposed solution of fp equation using XXXXX trained with XXXXX ga-sqp', ['neural networks', 'hybrid approach'])\n", "('NONE', 'XXXXX has been solved using XXXXX', ['optimization problem', 'genetic algorithms'])\n", "('method used for task', 'the XXXXX is equipped with a mixed XXXXX for the gait learning', ['kernel function', 'svm controller'])\n", "('method used for task', 'the XXXXX is trained based on XXXXX sizes', ['small sample', 'svm controller'])\n", "('method used for task', 'a novel XXXXX based algorithm inspired by superposition principle and field attraction for XXXXX', ['swarm intelligence', 'global optimization'])\n", "('method used for task', 'the XXXXX and XXXXX of the proposed method are increased by elimination of redundant features by using different feature selection methods', ['generalization capability', 'detection accuracy'])\n", "('NONE', 'the XXXXX and detection accuracy of the proposed method are increased by elimination of redundant features by using different XXXXX', ['generalization capability', 'feature selection methods'])\n", "('NONE', 'the generalization capability and XXXXX of the proposed method are increased by elimination of redundant features by using different XXXXX', ['detection accuracy', 'feature selection methods'])\n", "('NONE', 'our evolutionary XXXXX provides improved XXXXX', ['recognition rate', 'face recognition algorithm'])\n", "('NONE', 'XXXXX of XXXXX has been assessed', ['sensitivity analysis', 'decision maker preferences'])\n", "('method used for task', 'the balance factor is designed to maintain the diversity of solutions . the XXXXX is utilized to enhance the searching ability in entire XXXXX', ['scale factor', 'solution space'])\n", "('method used for task', 'the proposed algorithm is efficient for solving the large-scale XXXXX , especially for problems that have magnitude difference between two XXXXX', ['scheduling problems', 'optimization objectives'])\n", "('method used for task', 'this paper proposed a binary version of the XXXXX for solving 0-1 XXXXX', ['knapsack problem', 'monkey algorithm'])\n", "('method used for task', 'the ant system algorithm and XXXXX were combined to solve 3d/2d fixed-outline XXXXX', ['simulated annealing', 'floorplanning problem'])\n", "('NONE', 'a novel XXXXX using multiple XXXXX is designed', ['classification model', 'base classifiers'])\n", "('NONE', 'iapso outperforms several recent XXXXX , in terms of accuracy and XXXXX', ['meta-heuristic algorithms', 'convergence speed'])\n", "('method used for task', 'the concepts of XXXXX , route planning , and XXXXX are combined', ['fuzzy modeling', 'parallel machine scheduling'])\n", "('method used for task', 'the lower level problem is modeled as a mmbm problem , and a hungarian-based XXXXX ( hlp ) method is proposed , which can solve mmbm in XXXXX', ['linear programming', 'polynomial time'])\n", "('method used for task', 'the lower XXXXX is modeled as a mmbm problem , and a hungarian-based XXXXX ( hlp ) method is proposed , which can solve mmbm in polynomial time', ['linear programming', 'level problem'])\n", "('method used for task', 'the lower XXXXX is modeled as a mmbm problem , and a hungarian-based linear programming ( hlp ) method is proposed , which can solve mmbm in XXXXX', ['polynomial time', 'level problem'])\n", "('method used for task', 'the upper level optimization problem is solved by evolutionary multiobjective optimization algorithms , where a greedy reassignment local search operator , capable of leveraging the XXXXX and information from XXXXX , is introduced to improve the efficiency of the algorithm', ['domain knowledge', 'problem instances'])\n", "('method used for task', 'an XXXXX is determined for segmentation of XXXXX', ['adaptive threshold', 'power quality signals'])\n", "('method used for task', 'the XXXXX are based on XXXXX and acquired in field', ['mathematical models', 'power quality signals'])\n", "('method used for task', 'the intersections between the XXXXX and the XXXXX detail curves determine the start and the end of the segments', ['adaptive threshold', 'wavelet transform'])\n", "('method used for task', 'we develop a XXXXX to solve the green XXXXX', ['vehicle routing problem', 'solution approach'])\n", "('NONE', 'the proposed system utilized XXXXX for indicating the degree of XXXXX', ['case-based reasoning', 'water quality'])\n", "('NONE', 'three XXXXX , four XXXXX', ['evolutionary algorithms', 'fitness functions'])\n", "('NONE', 'cost minimized in a mix-product three echelon XXXXX while maintaining XXXXX', ['supply chain', 'service level'])\n", "('NONE', 'XXXXX , XXXXX , revised simos procedure , analytical hierarchy process', ['linear regression', 'factor analysis'])\n", "('NONE', 'XXXXX , factor analysis , revised simos procedure , analytical XXXXX', ['linear regression', 'hierarchy process'])\n", "('NONE', 'linear regression , XXXXX , revised simos procedure , analytical XXXXX', ['factor analysis', 'hierarchy process'])\n", "('NONE', 'minimization of maximum XXXXX of all parts is considered as XXXXX', ['completion time', 'objective function'])\n", "('method used for task', 'the aggregated XXXXX was used to investigate the simultaneous effects of printing parameters on the XXXXX and porosity of scaffolds', ['artificial neural network', 'compressive strength'])\n", "('NONE', 'most existing approaches reduce the XXXXX of qos-aware XXXXX to a single-objective problem using scalarization', ['multi-objective optimization problem', 'web service composition'])\n", "('method used for task', 'XXXXX ( de ) algorithms – in particular gde3 – yields the best results on this specific problem for several scenarios , also having the lowest XXXXX', ['differential evolution', 'time complexity'])\n", "('method used for task', 'the XXXXX for XXXXX is a critical decision making problem', ['location selection', 'nuclear power plant'])\n", "('NONE', 'the information of the XXXXX is used for online tuning of feedback gains of the XXXXX', ['pid controllers', 'wavelet fuzzy neural network'])\n", "('method used for task', 'the XXXXX approach is used to optimize local and global modelling capability of XXXXX', ['t-s fuzzy model', 'weighting parameters'])\n", "('NONE', 'the weighting parameters approach is used to optimize local and global XXXXX of XXXXX', ['t-s fuzzy model', 'modelling capability'])\n", "('method used for task', 'the XXXXX approach is used to optimize local and global XXXXX of t-s fuzzy model', ['weighting parameters', 'modelling capability'])\n", "('method used for task', 'a new XXXXX is developed for estimating the process change point in x¯ XXXXX', ['hybrid method', 'control chart'])\n", "('method used for task', 'a new XXXXX is developed for estimating the XXXXX in x¯ control chart', ['hybrid method', 'process change point'])\n", "('NONE', 'a new hybrid method is developed for estimating the XXXXX in x¯ XXXXX', ['control chart', 'process change point'])\n", "('NONE', 'the proposed XXXXX provides a more accurate estimate of the XXXXX', ['hybrid method', 'process change point'])\n", "('NONE', 'alc-pso is applied for the solution of XXXXX of XXXXX', ['power systems', 'opf problem'])\n", "('method used for task', 'a novel method for combination of XXXXX and fuzzy XXXXX ( frl ) is proposed', ['supervised learning', 'reinforcement learning'])\n", "('NONE', 'XXXXX is used for initialization of value ( worth ) of each candidate action of XXXXX in critic-only based frl algorithms', ['supervised learning', 'fuzzy rules'])\n", "('NONE', 'since the mamdani engines perform XXXXX on disjoint subsets of the XXXXX , the total number of the fuzzy rules needed for input–output mapping is far smaller', ['fuzzy reasoning', 'linguistic variables'])\n", "('NONE', 'since the mamdani engines perform XXXXX on disjoint subsets of the linguistic variables , the total number of the XXXXX needed for input–output mapping is far smaller', ['fuzzy reasoning', 'fuzzy rules'])\n", "('NONE', 'since the mamdani engines perform fuzzy reasoning on disjoint subsets of the XXXXX , the total number of the XXXXX needed for input–output mapping is far smaller', ['linguistic variables', 'fuzzy rules'])\n", "('method used for task', 'coupling the chain classification with greedy breadth-first-search ( gbfs ) algorithm , a new approach has been proposed for automated sizing of defects , utilizing XXXXX ( rbfnn ) and XXXXX ( svm ) . it has been established that the proposed approach can successfully size subsurface as well as surface defects', ['radial basis function neural network', 'support vector machine'])\n", "('NONE', 'coupling the chain classification with greedy breadth-first-search ( gbfs ) algorithm , a new approach has been proposed for automated sizing of defects , utilizing XXXXX ( rbfnn ) and support vector machine ( svm ) . it has been established that the proposed approach can successfully size subsurface as well as XXXXX', ['radial basis function neural network', 'surface defects'])\n", "('NONE', 'coupling the XXXXX with greedy breadth-first-search ( gbfs ) algorithm , a new approach has been proposed for automated sizing of defects , utilizing XXXXX ( rbfnn ) and support vector machine ( svm ) . it has been established that the proposed approach can successfully size subsurface as well as surface defects', ['radial basis function neural network', 'chain classification'])\n", "('NONE', 'coupling the chain classification with greedy breadth-first-search ( gbfs ) algorithm , a new approach has been proposed for automated sizing of defects , utilizing radial basis function neural network ( rbfnn ) and XXXXX ( svm ) . it has been established that the proposed approach can successfully size subsurface as well as XXXXX', ['support vector machine', 'surface defects'])\n", "('method used for task', 'coupling the XXXXX with greedy breadth-first-search ( gbfs ) algorithm , a new approach has been proposed for automated sizing of defects , utilizing radial basis function neural network ( rbfnn ) and XXXXX ( svm ) . it has been established that the proposed approach can successfully size subsurface as well as surface defects', ['support vector machine', 'chain classification'])\n", "('NONE', 'coupling the XXXXX with greedy breadth-first-search ( gbfs ) algorithm , a new approach has been proposed for automated sizing of defects , utilizing radial basis function neural network ( rbfnn ) and support vector machine ( svm ) . it has been established that the proposed approach can successfully size subsurface as well as XXXXX', ['surface defects', 'chain classification'])\n", "('NONE', 'XXXXX has incorporated dependency among the XXXXX ( length , width , depth and height ) and optimal sequence for sizing has been determined , for the first time', ['chain classification', 'class variables'])\n", "('method used for task', 'XXXXX has been able to significantly incorporate the dependency existing among the XXXXX and has successfully resulted in sizing of defects located even 3mm below the surface', ['chain classification', 'class variables'])\n", "('method used for task', 'odg problem is studied with an objective of reducing XXXXX and XXXXX', ['power loss', 'energy cost'])\n", "('NONE', 'eligible cluster heads are selected in the networks using XXXXX which ensures considerable XXXXX in the network', ['fuzzy logic', 'energy saving'])\n", "('method used for task', 'the nodes having highest number of neighbors , higher XXXXX and close to XXXXX are given preference in cluster head election', ['residual energy', 'base station'])\n", "('method used for task', 'the proposed method is applied on uci XXXXX repository datasets and its corresponding XXXXX are obtained', ['machine learning', 'performance measures'])\n", "('NONE', 'the comparison of the proposed XXXXX with the existing results reveal that the accuracy obtained by the present technique provides better XXXXX', ['similarity measures', 'similarity rating'])\n", "('method used for task', 'it is also observed that fixing the XXXXX for the XXXXX is difficult due to the negative rating in some existing methodology', ['similarity measures', 'threshold values'])\n", "('method used for task', 'the XXXXX obtained by the suggested technique is compared with the XXXXX data set and are analyzed', ['similarity rating', 'class distribution'])\n", "('method used for task', 'dynamic XXXXX model for obtaining the candidate XXXXX', ['expectation efficiency', 'objective function value'])\n", "('method used for task', 'static XXXXX model for updating the XXXXX', ['expectation efficiency', 'objective function value'])\n", "('NONE', 'simultaneous estimation of XXXXX and pixel labels of XXXXX is focus of work', ['bias field', 'mr images'])\n", "('method used for task', 'constrained XXXXX is used to tune the parameters of a robust XXXXX', ['particle swarm optimization algorithm', 'pid controller'])\n", "('NONE', 'XXXXX within the intervals of the XXXXX is ensured', ['robust stability', 'uncertain parameters'])\n", "('NONE', 'the method provides both a dynamic detection and fuzzy querying of XXXXX in crime-related XXXXX', ['time series', 'change points'])\n", "('NONE', 'XXXXX expressed with XXXXX can be used to query change points', ['geometric properties', 'linguistic variables'])\n", "('method used for task', 'XXXXX expressed with linguistic variables can be used to query XXXXX', ['geometric properties', 'change points'])\n", "('method used for task', 'geometric properties expressed with XXXXX can be used to query XXXXX', ['linguistic variables', 'change points'])\n", "('method used for task', 'we study a XXXXX system for outsourcing XXXXX ( orl )', ['reverse logistics', 'criteria evaluation'])\n", "('method used for task', 'analytic XXXXX is used to find the best XXXXX', ['compromise solution', 'hierarchy process'])\n", "('method used for task', 'this paper presents a new XXXXX for XXXXX with coa', ['data clustering', 'optimization approach'])\n", "('NONE', 'this paper presents a new optimization approach for XXXXX with coa and XXXXX for clustering data', ['data clustering', 'fuzzy system'])\n", "('method used for task', 'this paper presents a new XXXXX for XXXXX with coa and fuzzy system for clustering data', ['data clustering', 'optimization approach'])\n", "('method used for task', 'this paper presents a new XXXXX for data clustering with coa and XXXXX for clustering data', ['fuzzy system', 'optimization approach'])\n", "('NONE', 'an efficient proposal for XXXXX by cuckoo XXXXX', ['data clustering', 'optimization algorithm'])\n", "('NONE', 'all XXXXX are implemented using ds1104 dsp-based XXXXX', ['control algorithms', 'control computer'])\n", "('NONE', 'selection method based on a XXXXX optimized by a XXXXX', ['fuzzy system', 'genetic algorithm'])\n", "('method used for task', 'XXXXX based on a XXXXX optimized by a genetic algorithm', ['fuzzy system', 'selection method'])\n", "('method used for task', 'XXXXX based on a fuzzy system optimized by a XXXXX', ['genetic algorithm', 'selection method'])\n", "('NONE', 'the XXXXX of the XXXXX is designed with another fuzzy system in order to accomplish it optimization task', ['fitness function', 'genetic algorithm'])\n", "('NONE', 'the XXXXX of the genetic algorithm is designed with another XXXXX in order to accomplish it optimization task', ['fitness function', 'fuzzy system'])\n", "('method used for task', 'the fitness function of the XXXXX is designed with another XXXXX in order to accomplish it optimization task', ['genetic algorithm', 'fuzzy system'])\n", "('NONE', 'the neuro-fuzzy XXXXX uses sugeno type fuzzy rules with a randomized fuzzy layer and a linear XXXXX output layer', ['inference system', 'neural network'])\n", "('NONE', 'a novel XXXXX that makes a lot of sense when one has an absolute maximum number of XXXXX allowed in your application', ['performance evaluation method', 'function evaluations'])\n", "('method used for task', 'the XXXXX and precision of conventional pso is improved by introducing an adaptive inertia weight strategy based on the XXXXX of the particles', ['convergence speed', 'success rate'])\n", "('method used for task', 'a novel XXXXX for XXXXX under uncertainty is presented', ['soft computing approach', 'location decision'])\n", "('method used for task', 'the XXXXX is introduced for the parameter hmcr to improve the performance of the proposed XXXXX', ['dynamic adaptation', 'routing algorithm'])\n", "('NONE', 'an effective local search strategy is proposed to improve the XXXXX and the accuracy of the proposed XXXXX', ['convergence speed', 'routing algorithm'])\n", "('method used for task', 'an effective XXXXX is proposed to improve the XXXXX and the accuracy of the proposed routing algorithm', ['convergence speed', 'local search strategy'])\n", "('method used for task', 'an effective XXXXX is proposed to improve the convergence speed and the accuracy of the proposed XXXXX', ['routing algorithm', 'local search strategy'])\n", "('method used for task', 'XXXXX is trained to predict the XXXXX of composites from experimental variables', ['neural network', 'compressive strength'])\n", "('NONE', 'input variables are optimized to maximize XXXXX , by XXXXX', ['compressive strength', 'genetic algorithm'])\n", "('method used for task', 'XXXXX are optimized to maximize XXXXX , by genetic algorithm', ['compressive strength', 'input variables'])\n", "('method used for task', 'XXXXX are optimized to maximize compressive strength , by XXXXX', ['genetic algorithm', 'input variables'])\n", "('NONE', 'a robust neural-network-method applied to XXXXX in line-connected three-phase XXXXX is presented', ['speed estimation', 'induction motors'])\n", "('method used for task', 'a XXXXX of single and multiple current sensor applied to XXXXX is presented', ['comparative study', 'speed estimation'])\n", "('method used for task', 'we propose multiobjective XXXXX for XXXXX', ['evolutionary computation', 'multiple sequence alignment'])\n", "('NONE', 'unrelated parallel machines scheduling problem with past-sequence-dependent XXXXX , XXXXX , deteriorating jobs and learning effects is presented', ['setup times', 'release dates'])\n", "('NONE', 'unrelated parallel machines scheduling problem with past-sequence-dependent XXXXX , release dates , XXXXX and learning effects is presented', ['setup times', 'deteriorating jobs'])\n", "('method used for task', 'unrelated parallel machines scheduling problem with past-sequence-dependent XXXXX , release dates , deteriorating jobs and XXXXX is presented', ['setup times', 'learning effects'])\n", "('NONE', 'unrelated parallel machines scheduling problem with past-sequence-dependent setup times , XXXXX , XXXXX and learning effects is presented', ['release dates', 'deteriorating jobs'])\n", "('method used for task', 'unrelated parallel machines scheduling problem with past-sequence-dependent setup times , XXXXX , deteriorating jobs and XXXXX is presented', ['release dates', 'learning effects'])\n", "('method used for task', 'unrelated parallel machines scheduling problem with past-sequence-dependent setup times , release dates , XXXXX and XXXXX is presented', ['deteriorating jobs', 'learning effects'])\n", "('method used for task', 'XXXXX takes least estimation time and surpasses all other 11 XXXXX', ['firefly algorithm', 'metaheuristic algorithms'])\n", "('method used for task', 'results are compared with the results obtained using XXXXX and artificial bee XXXXX and it is proved that the proposed method offers reduced thd with less computation period', ['particle swarm optimization', 'colony algorithm'])\n", "('NONE', 'we studied the effect of XXXXX on multi-objective XXXXX', ['local search', 'memetic algorithm'])\n", "('NONE', 'morris’ oat method drives the XXXXX of the XXXXX', ['neighborhood search', 'abc algorithm'])\n", "('method used for task', \"the XXXXX is developed in nvidia 's cuda/c to reduce the XXXXX\", ['execution time', 'identification algorithm'])\n", "('method used for task', 'pfs models are compared with takagi-sugeno XXXXX and XXXXX', ['fuzzy models', 'logistic regression models'])\n", "('NONE', 'the application of using XXXXX in the XXXXX can be adopted suitably . all of members were able to assess dependently without any bias from the team members', ['fuzzy fmea', 'emergency department'])\n", "('method used for task', 'XXXXX can be applied for the first time to improve XXXXX in an emergency department of a public hospital', ['fuzzy fmea', 'decision making process'])\n", "('method used for task', 'XXXXX can be applied for the first time to improve decision making process in an XXXXX of a public hospital', ['fuzzy fmea', 'emergency department'])\n", "('NONE', 'fuzzy fmea can be applied for the first time to improve XXXXX in an XXXXX of a public hospital', ['decision making process', 'emergency department'])\n", "('NONE', 'the XXXXX are tuned by XXXXX', ['control parameters', 'fuzzy logic rules'])\n", "('NONE', 'the XXXXX was confirmed by XXXXX', ['performance improvement', 'simulation study'])\n", "('method used for task', 'XXXXX are carried out for XXXXX', ['computer simulations', 'performance evaluation'])\n", "('method used for task', 'conducting the XXXXX to understand in a better way the XXXXX', ['case study', 'evaluation process'])\n", "('method used for task', 'a XXXXX based on XXXXX is presented', ['binary particle swarm optimization', 'feature selection method'])\n", "('method used for task', 'fitness based adaptive XXXXX is integrated with the XXXXX to dynamically control the exploration and exploitation of the particle in the search space', ['inertia weight', 'binary particle swarm optimization'])\n", "('method used for task', 'fitness based adaptive XXXXX is integrated with the binary particle swarm optimization to dynamically control the exploration and exploitation of the particle in the XXXXX', ['inertia weight', 'search space'])\n", "('NONE', 'fitness based adaptive inertia weight is integrated with the XXXXX to dynamically control the exploration and exploitation of the particle in the XXXXX', ['binary particle swarm optimization', 'search space'])\n", "('method used for task', 'the kernel principal component analysis is used to transform nonlinear XXXXX to improve XXXXX', ['feature space', 'classification performance'])\n", "('method used for task', 'XXXXX , which is regarded as an important and practically useful XXXXX , is fully exploited for designing sugeno-type granular model', ['information granularity', 'design asset'])\n", "('NONE', 'the proposed approach of constructing the sugeno-type granular model is of general nature as it could be applied to various XXXXX and realized by invoking different formalisms of XXXXX', ['fuzzy models', 'information granules'])\n", "('method used for task', 'a XXXXX based on hesitant fuzzy XXXXX is proposed.it is capable of taking into account the fundamental building blocks of trust.it considers the hesitancy and vagueness of trust decision making process.it considers the subjectivity and uncertainty of trust decision making process.its effectiveness is demonstrated by providing several evaluation scenarios', ['trust model', 'multi-criteria decision making'])\n", "('method used for task', 'a XXXXX based on hesitant fuzzy multi-criteria decision making is proposed.it is capable of taking into account the fundamental XXXXX of trust.it considers the hesitancy and vagueness of trust decision making process.it considers the subjectivity and uncertainty of trust decision making process.its effectiveness is demonstrated by providing several evaluation scenarios', ['trust model', 'building blocks'])\n", "('method used for task', 'a XXXXX based on hesitant fuzzy multi-criteria XXXXX is proposed.it is capable of taking into account the fundamental building blocks of trust.it considers the hesitancy and vagueness of trust XXXXX process.it considers the subjectivity and uncertainty of trust XXXXX process.its effectiveness is demonstrated by providing several evaluation scenarios', ['trust model', 'decision making'])\n", "('method used for task', 'a XXXXX based on hesitant fuzzy multi-criteria decision making is proposed.it is capable of taking into account the fundamental building blocks of trust.it considers the hesitancy and vagueness of XXXXX process.it considers the subjectivity and uncertainty of XXXXX process.its effectiveness is demonstrated by providing several evaluation scenarios', ['trust model', 'trust decision making'])\n", "('method used for task', 'a XXXXX based on hesitant fuzzy multi-criteria decision making is proposed.it is capable of taking into account the fundamental building blocks of trust.it considers the hesitancy and vagueness of XXXXX process.it considers the subjectivity and uncertainty of XXXXX process.its effectiveness is demonstrated by providing several evaluation scenarios', ['trust model', 'trust decision making'])\n", "('method used for task', 'a trust model based on hesitant fuzzy XXXXX is proposed.it is capable of taking into account the fundamental XXXXX of trust.it considers the hesitancy and vagueness of trust decision making process.it considers the subjectivity and uncertainty of trust decision making process.its effectiveness is demonstrated by providing several evaluation scenarios', ['multi-criteria decision making', 'building blocks'])\n", "('NONE', 'a trust model based on hesitant fuzzy XXXXX is proposed.it is capable of taking into account the fundamental building blocks of trust.it considers the hesitancy and vagueness of trust XXXXX process.it considers the subjectivity and uncertainty of trust XXXXX process.its effectiveness is demonstrated by providing several evaluation scenarios', ['multi-criteria decision making', 'decision making'])\n", "('NONE', 'a trust model based on hesitant fuzzy XXXXX is proposed.it is capable of taking into account the fundamental building blocks of trust.it considers the hesitancy and vagueness of XXXXX process.it considers the subjectivity and uncertainty of XXXXX process.its effectiveness is demonstrated by providing several evaluation scenarios', ['multi-criteria decision making', 'trust decision making'])\n", "('NONE', 'a trust model based on hesitant fuzzy XXXXX is proposed.it is capable of taking into account the fundamental building blocks of trust.it considers the hesitancy and vagueness of XXXXX process.it considers the subjectivity and uncertainty of XXXXX process.its effectiveness is demonstrated by providing several evaluation scenarios', ['multi-criteria decision making', 'trust decision making'])\n", "('method used for task', 'a trust model based on hesitant fuzzy multi-criteria XXXXX is proposed.it is capable of taking into account the fundamental XXXXX of trust.it considers the hesitancy and vagueness of trust XXXXX process.it considers the subjectivity and uncertainty of trust XXXXX process.its effectiveness is demonstrated by providing several evaluation scenarios', ['building blocks', 'decision making'])\n", "('NONE', 'a trust model based on hesitant fuzzy multi-criteria decision making is proposed.it is capable of taking into account the fundamental XXXXX of trust.it considers the hesitancy and vagueness of XXXXX process.it considers the subjectivity and uncertainty of XXXXX process.its effectiveness is demonstrated by providing several evaluation scenarios', ['building blocks', 'trust decision making'])\n", "('NONE', 'a trust model based on hesitant fuzzy multi-criteria decision making is proposed.it is capable of taking into account the fundamental XXXXX of trust.it considers the hesitancy and vagueness of XXXXX process.it considers the subjectivity and uncertainty of XXXXX process.its effectiveness is demonstrated by providing several evaluation scenarios', ['building blocks', 'trust decision making'])\n", "('NONE', 'a trust model based on hesitant fuzzy multi-criteria XXXXX is proposed.it is capable of taking into account the fundamental building blocks of trust.it considers the hesitancy and vagueness of trust XXXXX process.it considers the subjectivity and uncertainty of trust XXXXX process.its effectiveness is demonstrated by providing several evaluation scenarios', ['decision making', 'trust decision making'])\n", "('NONE', 'a trust model based on hesitant fuzzy multi-criteria XXXXX is proposed.it is capable of taking into account the fundamental building blocks of trust.it considers the hesitancy and vagueness of trust XXXXX process.it considers the subjectivity and uncertainty of trust XXXXX process.its effectiveness is demonstrated by providing several evaluation scenarios', ['decision making', 'trust decision making'])\n", "('NONE', 'XXXXX shows that XXXXX is efficient to work with gmf', ['comparative analysis', 'differential evolution'])\n", "('NONE', 'we propose a XXXXX by propagating probabilities between XXXXX', ['clustering algorithm', 'data points'])\n", "('method used for task', 'develop new XXXXX for interval reciprocal XXXXX ( irprs )', ['similarity measures', 'preference relations'])\n", "('method used for task', 'a hybrid XXXXX based on XXXXX is proposed', ['particle swarm optimization', 'feature selection method'])\n", "('method used for task', 'our method uses a novel XXXXX to enhance the XXXXX near global optima', ['local search', 'search process'])\n", "('method used for task', 'natural XXXXX is used in XXXXX', ['evolution strategy', 'structural optimization'])\n", "('method used for task', 'using XXXXX for XXXXX and rule combination', ['genetic algorithm', 'parameter selection'])\n", "('NONE', 'applying hybrid XXXXX on real life very XXXXX', ['evolutionary methods', 'large data set'])\n", "('method used for task', 'to reduce the XXXXX , the modified cnfslms algorithm is proposed by replacing the XXXXX with a versorial function', ['computational complexity', 'sigmoid function'])\n", "('NONE', 'the idea treats XXXXX as an essential XXXXX', ['information granularity', 'design asset'])\n", "('method used for task', 'XXXXX based on XXXXX are completed', ['comparative studies', 'global sensitivity analysis'])\n", "('method used for task', 'parallel XXXXX ( pvns ) is implemented for solving bus terminal XXXXX ( btlp )', ['variable neighborhood search', 'location problem'])\n", "('method used for task', 'improved XXXXX based on the neighborhoods fast interchange is combined with reduced XXXXX based on the covering characteristic of the problem', ['local search', 'neighborhood size'])\n", "('NONE', 'we study the XXXXX of finding a XXXXX of graphs based on rough sets', ['np-hard problem', 'minimum vertex cover'])\n", "('NONE', 'we study the XXXXX of finding a minimum vertex cover of graphs based on XXXXX', ['np-hard problem', 'rough sets'])\n", "('NONE', 'we study the np-hard problem of finding a XXXXX of graphs based on XXXXX', ['minimum vertex cover', 'rough sets'])\n", "('NONE', 'the problem of finding a XXXXX of graphs can be translated into the problem of finding an optimal reduct of a decision information table in XXXXX', ['minimum vertex cover', 'rough sets'])\n", "('NONE', 'the problem of finding a XXXXX of graphs can be translated into the problem of finding an optimal reduct of a XXXXX table in rough sets', ['minimum vertex cover', 'decision information'])\n", "('NONE', 'the problem of finding a minimum vertex cover of graphs can be translated into the problem of finding an optimal reduct of a XXXXX table in XXXXX', ['rough sets', 'decision information'])\n", "('method used for task', 'a new method based on XXXXX is proposed to compute the XXXXX of a given graph', ['rough sets', 'minimum vertex cover'])\n", "('method used for task', 'presenting two algorithms for ivif XXXXX and XXXXX', ['incomplete information', 'importance weights'])\n", "('method used for task', 'XXXXX for the XXXXX parameters have been derived from the lyapunov–krasovskii functional candidate', ['adaptation laws', 'type-2 fuzzy logic system controller'])\n", "('method used for task', 'the congestion cost , rescheduled real power , XXXXX and XXXXX are less using hnm-fapso', ['power loss', 'computation time'])\n", "('NONE', 'the XXXXX , rescheduled real power , XXXXX and computation time are less using hnm-fapso', ['power loss', 'congestion cost'])\n", "('method used for task', 'the XXXXX , rescheduled real power , power loss and XXXXX are less using hnm-fapso', ['computation time', 'congestion cost'])\n", "('NONE', 'a multi-product multi-period XXXXX under XXXXX is developed', ['budget constraint', 'inventory control problem'])\n", "('NONE', 'a novel XXXXX selection algorithm that uses XXXXX for XXXXX selection', ['fuzzy logic', 'cluster head'])\n", "('NONE', 'a novel XXXXX algorithm that uses XXXXX for XXXXX', ['fuzzy logic', 'cluster head selection'])\n", "('NONE', 'a novel XXXXX algorithm that uses XXXXX for XXXXX', ['fuzzy logic', 'cluster head selection'])\n", "('method used for task', 'a novel XXXXX selection algorithm that uses fuzzy logic for XXXXX selection', ['cluster head', 'cluster head selection'])\n", "('method used for task', 'a novel XXXXX selection algorithm that uses fuzzy logic for XXXXX selection', ['cluster head', 'cluster head selection'])\n", "('method used for task', 'XXXXX is used to ensemble the XXXXX', ['weighted voting mechanism', 'base classifiers'])\n", "('method used for task', 'we propose a XXXXX model based on fuzzy XXXXX', ['logistic regression', 'technology credit scoring'])\n", "('method used for task', 'a novel XXXXX to forecast XXXXX prices was built', ['neuro-fuzzy controller', 'carbon emission'])\n", "('NONE', 'we have designed a XXXXX embedded device that would reduce the penalty billing of the consumers who fail to maintain the desired XXXXX', ['low cost', 'power factor'])\n", "('method used for task', 'switching of XXXXX banks though economical but often fail to give satisfactory compensation for varying load patterns . hence combination of XXXXX banks intelligently would give a better solution to this problem', ['shunt capacitor', 'shunt banks'])\n", "('method used for task', 'switching of XXXXX banks though economical but often fail to give satisfactory compensation for varying XXXXX . hence combination of XXXXX banks intelligently would give a better solution to this problem', ['shunt capacitor', 'load patterns'])\n", "('method used for task', 'switching of XXXXX banks though economical but often fail to give satisfactory compensation for varying load patterns . hence combination of XXXXX banks intelligently would give a better solution to this problem', ['shunt capacitor', 'shunt banks'])\n", "('method used for task', 'switching of XXXXX banks though economical but often fail to give satisfactory compensation for varying load patterns . hence combination of XXXXX banks intelligently would give a better solution to this problem', ['shunt banks', 'shunt capacitor'])\n", "('method used for task', 'switching of XXXXX banks though economical but often fail to give satisfactory compensation for varying XXXXX . hence combination of XXXXX banks intelligently would give a better solution to this problem', ['load patterns', 'shunt capacitor'])\n", "('method used for task', 'switching of XXXXX banks though economical but often fail to give satisfactory compensation for varying load patterns . hence combination of XXXXX banks intelligently would give a better solution to this problem', ['shunt capacitor', 'shunt banks'])\n", "('method used for task', 'we are presenting a modification of a XXXXX based on the bee behavior for optimizing XXXXX', ['bio-inspired algorithm', 'fuzzy controllers'])\n", "('NONE', 'XXXXX with XXXXX are used as benchmarks', ['vehicle routing problems', 'time windows'])\n", "('NONE', 'it starts the XXXXX with an XXXXX consisting of viruses and host cells . then , the environment will be clustered to number of regions called virus groups', ['initial population', 'optimization process'])\n", "('NONE', 'comparing the result of this algorithm with other well-known XXXXX demonstrates the superiority of the proposed algorithm in terms of the global solution convergence and the XXXXX', ['optimization algorithms', 'convergence speed'])\n", "('NONE', 'a two-level self-adaptive XXXXX is developed based on the two levels of decisions in the pcvrp , namely the selection of customers and the XXXXX of selected customers', ['vehicle scheduling', 'variable neighborhood search algorithm'])\n", "('method used for task', 'a graph extension method is adopted to obtain the XXXXX for pcvrp by transforming the proposed XXXXX of pcvrp into an equivalent traveling salesman problem model', ['lower bound', 'mixed integer programming model'])\n", "('NONE', 'XXXXX of XXXXX using legendre polynomial based functional link artificial neural network ( flann )', ['numerical solution', 'ordinary differential equations'])\n", "('NONE', 'the selection of XXXXX of ema is done through exhaustive XXXXX', ['control parameters', 'parametric study'])\n", "('method used for task', 'propose a new XXXXX to select best XXXXX', ['pareto front', 'decision making method'])\n", "('method used for task', 'in the present study , an XXXXX was used to solve a discrete mcr problem . during the solving process , the social force model was used to improve the XXXXX , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the XXXXX revealed that our algorithm was superior to the exact algorithm and greedy algorithm in terms of solution quality and running time', ['ant colony algorithm', 'simulation experiments'])\n", "('method used for task', 'in the present study , an XXXXX was used to solve a discrete mcr problem . during the solving process , the social force model was used to improve the XXXXX , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the simulation experiments revealed that our algorithm was superior to the XXXXX and greedy algorithm in terms of solution quality and running time', ['ant colony algorithm', 'exact algorithm'])\n", "('method used for task', 'in the present study , an XXXXX was used to solve a discrete mcr problem . during the solving process , the social force model was used to improve the XXXXX , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the simulation experiments revealed that our algorithm was superior to the exact algorithm and XXXXX in terms of solution quality and running time', ['ant colony algorithm', 'greedy algorithm'])\n", "('method used for task', 'in the present study , an XXXXX was used to solve a discrete mcr problem . during the solving process , the social force model was used to improve the XXXXX , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the simulation experiments revealed that our algorithm was superior to the exact algorithm and greedy algorithm in terms of solution quality and XXXXX', ['ant colony algorithm', 'running time'])\n", "('method used for task', 'in the present study , an XXXXX was used to solve a discrete mcr problem . during the solving process , the social XXXXX was used to improve the XXXXX , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the simulation experiments revealed that our algorithm was superior to the exact algorithm and greedy algorithm in terms of solution quality and running time', ['ant colony algorithm', 'force model'])\n", "('method used for task', 'in the present study , an XXXXX was used to solve a discrete mcr problem . during the solving process , the social force model was used to improve the XXXXX , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the simulation experiments revealed that our algorithm was superior to the exact algorithm and greedy algorithm in terms of XXXXX and running time', ['ant colony algorithm', 'solution quality'])\n", "('method used for task', 'in the present study , an XXXXX was used to solve a discrete mcr problem . during the solving process , the social force model was used to improve the XXXXX , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the XXXXX revealed that our algorithm was superior to the exact algorithm and greedy algorithm in terms of solution quality and running time', ['ant colony algorithm', 'simulation experiments'])\n", "('method used for task', 'in the present study , an XXXXX was used to solve a discrete mcr problem . during the solving process , the social force model was used to improve the XXXXX , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the simulation experiments revealed that our algorithm was superior to the XXXXX and greedy algorithm in terms of solution quality and running time', ['ant colony algorithm', 'exact algorithm'])\n", "('method used for task', 'in the present study , an XXXXX was used to solve a discrete mcr problem . during the solving process , the social force model was used to improve the XXXXX , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the simulation experiments revealed that our algorithm was superior to the exact algorithm and XXXXX in terms of solution quality and running time', ['ant colony algorithm', 'greedy algorithm'])\n", "('method used for task', 'in the present study , an XXXXX was used to solve a discrete mcr problem . during the solving process , the social force model was used to improve the XXXXX , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the simulation experiments revealed that our algorithm was superior to the exact algorithm and greedy algorithm in terms of solution quality and XXXXX', ['ant colony algorithm', 'running time'])\n", "('method used for task', 'in the present study , an XXXXX was used to solve a discrete mcr problem . during the solving process , the social XXXXX was used to improve the XXXXX , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the simulation experiments revealed that our algorithm was superior to the exact algorithm and greedy algorithm in terms of solution quality and running time', ['ant colony algorithm', 'force model'])\n", "('method used for task', 'in the present study , an XXXXX was used to solve a discrete mcr problem . during the solving process , the social force model was used to improve the XXXXX , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the simulation experiments revealed that our algorithm was superior to the exact algorithm and greedy algorithm in terms of XXXXX and running time', ['ant colony algorithm', 'solution quality'])\n", "('method used for task', 'in the present study , an ant colony algorithm was used to solve a discrete mcr problem . during the solving process , the social force model was used to improve the ant colony algorithm , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the XXXXX revealed that our algorithm was superior to the XXXXX and greedy algorithm in terms of solution quality and running time', ['simulation experiments', 'exact algorithm'])\n", "('method used for task', 'in the present study , an ant colony algorithm was used to solve a discrete mcr problem . during the solving process , the social force model was used to improve the ant colony algorithm , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the XXXXX revealed that our algorithm was superior to the exact algorithm and XXXXX in terms of solution quality and running time', ['simulation experiments', 'greedy algorithm'])\n", "('NONE', 'in the present study , an ant colony algorithm was used to solve a discrete mcr problem . during the solving process , the social force model was used to improve the ant colony algorithm , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the XXXXX revealed that our algorithm was superior to the exact algorithm and greedy algorithm in terms of solution quality and XXXXX', ['simulation experiments', 'running time'])\n", "('method used for task', 'in the present study , an ant colony algorithm was used to solve a discrete mcr problem . during the solving process , the social XXXXX was used to improve the ant colony algorithm , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the XXXXX revealed that our algorithm was superior to the exact algorithm and greedy algorithm in terms of solution quality and running time', ['simulation experiments', 'force model'])\n", "('NONE', 'in the present study , an ant colony algorithm was used to solve a discrete mcr problem . during the solving process , the social force model was used to improve the ant colony algorithm , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the XXXXX revealed that our algorithm was superior to the exact algorithm and greedy algorithm in terms of XXXXX and running time', ['simulation experiments', 'solution quality'])\n", "('method used for task', 'in the present study , an ant colony algorithm was used to solve a discrete mcr problem . during the solving process , the social force model was used to improve the ant colony algorithm , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the simulation experiments revealed that our algorithm was superior to the XXXXX and XXXXX in terms of solution quality and running time', ['exact algorithm', 'greedy algorithm'])\n", "('NONE', 'in the present study , an ant colony algorithm was used to solve a discrete mcr problem . during the solving process , the social force model was used to improve the ant colony algorithm , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the simulation experiments revealed that our algorithm was superior to the XXXXX and greedy algorithm in terms of solution quality and XXXXX', ['exact algorithm', 'running time'])\n", "('method used for task', 'in the present study , an ant colony algorithm was used to solve a discrete mcr problem . during the solving process , the social XXXXX was used to improve the ant colony algorithm , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the simulation experiments revealed that our algorithm was superior to the XXXXX and greedy algorithm in terms of solution quality and running time', ['exact algorithm', 'force model'])\n", "('NONE', 'in the present study , an ant colony algorithm was used to solve a discrete mcr problem . during the solving process , the social force model was used to improve the ant colony algorithm , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the simulation experiments revealed that our algorithm was superior to the XXXXX and greedy algorithm in terms of XXXXX and running time', ['exact algorithm', 'solution quality'])\n", "('NONE', 'in the present study , an ant colony algorithm was used to solve a discrete mcr problem . during the solving process , the social force model was used to improve the ant colony algorithm , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the simulation experiments revealed that our algorithm was superior to the exact algorithm and XXXXX in terms of solution quality and XXXXX', ['greedy algorithm', 'running time'])\n", "('method used for task', 'in the present study , an ant colony algorithm was used to solve a discrete mcr problem . during the solving process , the social XXXXX was used to improve the ant colony algorithm , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the simulation experiments revealed that our algorithm was superior to the exact algorithm and XXXXX in terms of solution quality and running time', ['greedy algorithm', 'force model'])\n", "('NONE', 'in the present study , an ant colony algorithm was used to solve a discrete mcr problem . during the solving process , the social force model was used to improve the ant colony algorithm , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the simulation experiments revealed that our algorithm was superior to the exact algorithm and XXXXX in terms of XXXXX and running time', ['greedy algorithm', 'solution quality'])\n", "('method used for task', 'in the present study , an ant colony algorithm was used to solve a discrete mcr problem . during the solving process , the social XXXXX was used to improve the ant colony algorithm , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the simulation experiments revealed that our algorithm was superior to the exact algorithm and greedy algorithm in terms of solution quality and XXXXX', ['running time', 'force model'])\n", "('method used for task', 'in the present study , an ant colony algorithm was used to solve a discrete mcr problem . during the solving process , the social force model was used to improve the ant colony algorithm , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the simulation experiments revealed that our algorithm was superior to the exact algorithm and greedy algorithm in terms of XXXXX and XXXXX', ['running time', 'solution quality'])\n", "('method used for task', 'in the present study , an ant colony algorithm was used to solve a discrete mcr problem . during the solving process , the social XXXXX was used to improve the ant colony algorithm , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the simulation experiments revealed that our algorithm was superior to the exact algorithm and greedy algorithm in terms of XXXXX and running time', ['force model', 'solution quality'])\n", "('method used for task', 'scd is predicated using XXXXX and sudden XXXXX index ( scdi )', ['svm classifier', 'cardiac death'])\n", "('NONE', 'XXXXX are extracted from XXXXX', ['nonlinear features', 'hrv signals'])\n", "('method used for task', 'a novel de-noising method is proposed to remove the ecg interference from noisy emgdi signals . experimental results achieved on practical XXXXX show that the proposed approach is better than traditional methods include XXXXX ( wt ) , ica , digital filter and adaptive filter in ecg interference removing', ['clinical data', 'wavelet transform'])\n", "('method used for task', 'a novel de-noising method is proposed to remove the ecg interference from noisy emgdi signals . experimental results achieved on practical XXXXX show that the proposed approach is better than traditional methods include wavelet transform ( wt ) , ica , XXXXX and adaptive filter in ecg interference removing', ['clinical data', 'digital filter'])\n", "('NONE', 'a novel de-noising method is proposed to remove the ecg interference from noisy emgdi signals . experimental results achieved on practical clinical data show that the proposed approach is better than traditional methods include XXXXX ( wt ) , ica , XXXXX and adaptive filter in ecg interference removing', ['wavelet transform', 'digital filter'])\n", "('NONE', 'the original XXXXX of contaminated emgdi signal were firstly obtained with independent XXXXX ( ica ) . then the ecg components contained were removed by a specially designed wavelet domain filter . finally , the purified XXXXX were reconstructed back to the original signal space by ica to obtain the clean emgdi signals', ['independent components', 'component analysis'])\n", "('NONE', 'the original XXXXX of contaminated emgdi signal were firstly obtained with independent component analysis ( ica ) . then the ecg components contained were removed by a specially designed XXXXX filter . finally , the purified XXXXX were reconstructed back to the original signal space by ica to obtain the clean emgdi signals', ['independent components', 'wavelet domain'])\n", "('NONE', 'the original XXXXX of contaminated emgdi signal were firstly obtained with independent XXXXX ( ica ) . then the ecg components contained were removed by a specially designed wavelet domain filter . finally , the purified XXXXX were reconstructed back to the original signal space by ica to obtain the clean emgdi signals', ['independent components', 'component analysis'])\n", "('NONE', 'the original XXXXX of contaminated emgdi signal were firstly obtained with independent component analysis ( ica ) . then the ecg components contained were removed by a specially designed XXXXX filter . finally , the purified XXXXX were reconstructed back to the original signal space by ica to obtain the clean emgdi signals', ['independent components', 'wavelet domain'])\n", "('NONE', 'the original independent components of contaminated emgdi signal were firstly obtained with independent XXXXX ( ica ) . then the ecg components contained were removed by a specially designed XXXXX filter . finally , the purified independent components were reconstructed back to the original signal space by ica to obtain the clean emgdi signals', ['component analysis', 'wavelet domain'])\n", "('method used for task', 'the classification is performed using two different classifiers ; XXXXX and XXXXX', ['support vector machine', 'k-nearest neighbors'])\n", "('NONE', 'the actual plant input is known to the controller at each XXXXX , therefore , it can cope with relatively large network delays and XXXXX', ['time step', 'packet losses'])\n", "('NONE', 'shown improved XXXXX using battery as a XXXXX', ['prediction accuracy', 'case study'])\n", "('NONE', 'we propose a XXXXX for elm using XXXXX ( ga )', ['genetic algorithms', 'pruning method'])\n", "('NONE', 'this paper proposes a new XXXXX based on rules of XXXXX defined within the framework of cell-like p-systems to solve sudoku', ['particle swarm optimization', 'membrane algorithm'])\n", "('method used for task', 'also , sudoku of XXXXX is solved for which the XXXXX is about 87 %', ['higher order', 'success rate'])\n", "('NONE', 'obtain good XXXXX in terms of the solution quality and XXXXX', ['convergence speed', 'reconstruction performance'])\n", "('method used for task', 'obtain good reconstruction performance in terms of the XXXXX and XXXXX', ['convergence speed', 'solution quality'])\n", "('NONE', 'obtain good XXXXX in terms of the XXXXX and convergence speed', ['reconstruction performance', 'solution quality'])\n", "('method used for task', 'the novel XXXXX for the XXXXX', ['genetic algorithm', 'protein structure prediction problem'])\n", "('NONE', 'our XXXXX is developed by the adoption of the XXXXX', ['incremental algorithm', 'attribute reduction process'])\n", "('NONE', 'this paper presents an approach for XXXXX in digital mammograms using XXXXX', ['breast cancer diagnosis', 'shearlet transform'])\n", "('NONE', 'a new approach is represented for XXXXX using XXXXX , t-test statistic and dynamic thresholding', ['feature extraction', 'shearlet transform'])\n", "('NONE', 'we analyze the behavior of this XXXXX under the presence of XXXXX', ['neural network', 'class imbalance'])\n", "('method used for task', 'a heuristic ant colony optimization approach with adjustment strategy is proposed for the bearing surface packing problem so that the XXXXX is reduced from real space to integer space and the XXXXX is improved obviously', ['search space', 'computational efficiency'])\n", "('NONE', 'a XXXXX ( pso ) approach is designed to optimize the mass center and inertia angles of the satellite module in a way of rotation so that the XXXXX is improved significantly', ['particle swarm optimization', 'solution quality'])\n", "('method used for task', 'the solution quality of the proposed hybrid multi-mechanism optimization approach ( hmmoa ) is better than those of the dual-system variable-grain cooperative co-evolutionary XXXXX ( dvgccga ) and XXXXX through an organism combination of the quasi-principal component analysis , XXXXX and pso ( qpgp )', ['genetic algorithm', 'hybrid method'])\n", "('NONE', 'the XXXXX of the proposed hybrid multi-mechanism optimization approach ( hmmoa ) is better than those of the dual-system variable-grain cooperative co-evolutionary XXXXX ( dvgccga ) and hybrid method through an organism combination of the quasi-principal component analysis , XXXXX and pso ( qpgp )', ['genetic algorithm', 'solution quality'])\n", "('NONE', 'the solution quality of the proposed hybrid multi-mechanism XXXXX ( hmmoa ) is better than those of the dual-system variable-grain cooperative co-evolutionary XXXXX ( dvgccga ) and hybrid method through an organism combination of the quasi-principal component analysis , XXXXX and pso ( qpgp )', ['genetic algorithm', 'optimization approach'])\n", "('NONE', 'the solution quality of the proposed hybrid multi-mechanism optimization approach ( hmmoa ) is better than those of the dual-system variable-grain cooperative co-evolutionary XXXXX ( dvgccga ) and hybrid method through an organism combination of the quasi-principal XXXXX , XXXXX and pso ( qpgp )', ['genetic algorithm', 'component analysis'])\n", "('method used for task', 'the solution quality of the proposed hybrid multi-mechanism optimization approach ( hmmoa ) is better than those of the dual-system variable-grain cooperative co-evolutionary XXXXX ( dvgccga ) and XXXXX through an organism combination of the quasi-principal component analysis , XXXXX and pso ( qpgp )', ['hybrid method', 'genetic algorithm'])\n", "('NONE', 'the XXXXX of the proposed hybrid multi-mechanism optimization approach ( hmmoa ) is better than those of the dual-system variable-grain cooperative co-evolutionary genetic algorithm ( dvgccga ) and XXXXX through an organism combination of the quasi-principal component analysis , genetic algorithm and pso ( qpgp )', ['hybrid method', 'solution quality'])\n", "('method used for task', 'the solution quality of the proposed hybrid multi-mechanism XXXXX ( hmmoa ) is better than those of the dual-system variable-grain cooperative co-evolutionary genetic algorithm ( dvgccga ) and XXXXX through an organism combination of the quasi-principal component analysis , genetic algorithm and pso ( qpgp )', ['hybrid method', 'optimization approach'])\n", "('NONE', 'the solution quality of the proposed hybrid multi-mechanism optimization approach ( hmmoa ) is better than those of the dual-system variable-grain cooperative co-evolutionary genetic algorithm ( dvgccga ) and XXXXX through an organism combination of the quasi-principal XXXXX , genetic algorithm and pso ( qpgp )', ['hybrid method', 'component analysis'])\n", "('NONE', 'the XXXXX of the proposed hybrid multi-mechanism optimization approach ( hmmoa ) is better than those of the dual-system variable-grain cooperative co-evolutionary XXXXX ( dvgccga ) and hybrid method through an organism combination of the quasi-principal component analysis , XXXXX and pso ( qpgp )', ['genetic algorithm', 'solution quality'])\n", "('NONE', 'the solution quality of the proposed hybrid multi-mechanism XXXXX ( hmmoa ) is better than those of the dual-system variable-grain cooperative co-evolutionary XXXXX ( dvgccga ) and hybrid method through an organism combination of the quasi-principal component analysis , XXXXX and pso ( qpgp )', ['genetic algorithm', 'optimization approach'])\n", "('NONE', 'the solution quality of the proposed hybrid multi-mechanism optimization approach ( hmmoa ) is better than those of the dual-system variable-grain cooperative co-evolutionary XXXXX ( dvgccga ) and hybrid method through an organism combination of the quasi-principal XXXXX , XXXXX and pso ( qpgp )', ['genetic algorithm', 'component analysis'])\n", "('NONE', 'the XXXXX of the proposed hybrid multi-mechanism XXXXX ( hmmoa ) is better than those of the dual-system variable-grain cooperative co-evolutionary genetic algorithm ( dvgccga ) and hybrid method through an organism combination of the quasi-principal component analysis , genetic algorithm and pso ( qpgp )', ['solution quality', 'optimization approach'])\n", "('NONE', 'the XXXXX of the proposed hybrid multi-mechanism optimization approach ( hmmoa ) is better than those of the dual-system variable-grain cooperative co-evolutionary genetic algorithm ( dvgccga ) and hybrid method through an organism combination of the quasi-principal XXXXX , genetic algorithm and pso ( qpgp )', ['solution quality', 'component analysis'])\n", "('NONE', 'the solution quality of the proposed hybrid multi-mechanism XXXXX ( hmmoa ) is better than those of the dual-system variable-grain cooperative co-evolutionary genetic algorithm ( dvgccga ) and hybrid method through an organism combination of the quasi-principal XXXXX , genetic algorithm and pso ( qpgp )', ['optimization approach', 'component analysis'])\n", "('NONE', 'a new ipd model of XXXXX with adaptive XXXXX was built', ['multiple agents', 'risk attitudes'])\n", "('method used for task', 'XXXXX and XXXXX provide the best intervals', ['genetic programming', 'linear regression'])\n", "('method used for task', 'this paper studies the XXXXX and XXXXX with class setups in a single machine environment', ['order acceptance', 'scheduling problem'])\n", "('NONE', 'this XXXXX the XXXXX and scheduling problem with class setups in a single machine environment', ['order acceptance', 'paper studies'])\n", "('method used for task', 'this XXXXX the order acceptance and XXXXX with class setups in a single machine environment', ['scheduling problem', 'paper studies'])\n", "('NONE', 'a hybrid XXXXX in a XXXXX is proposed to solve the aircraft landing problem', ['particle swarm optimization algorithm', 'rolling horizon framework'])\n", "('method used for task', 'the main contributions of this paper are a strategy based on pso and XXXXX for optimizing a XXXXX that responds stronger to multi-scale and multi-orientation texture micro-patterns in a mammogram and enhances the classification rate', ['incremental clustering', 'gabor filter bank'])\n", "('method used for task', 'the main contributions of this paper are a strategy based on pso and XXXXX for optimizing a gabor filter bank that responds stronger to multi-scale and multi-orientation texture micro-patterns in a mammogram and enhances the XXXXX', ['incremental clustering', 'classification rate'])\n", "('method used for task', 'the main contributions of this paper are a strategy based on pso and incremental clustering for optimizing a XXXXX that responds stronger to multi-scale and multi-orientation texture micro-patterns in a mammogram and enhances the XXXXX', ['gabor filter bank', 'classification rate'])\n", "('method used for task', 'for the evaluation of the effect of the optimized XXXXX on the mass classification problems , it is applied on overlapping blocks of the rois to collect moment-based features ( i.e. , mean , XXXXX , skewness ) from the magnitudes of gabor responses', ['gabor filter bank', 'standard deviation'])\n", "('NONE', 'we focus on XXXXX ( mc ) method in XXXXX', ['monte carlo', 'fuzzy linear regression models'])\n", "('NONE', 'the study covers different XXXXX that have not previously calculated for monte carlo study in XXXXX', ['error measures', 'fuzzy linear regression models'])\n", "('method used for task', 'we obtain the most useful and the worst XXXXX to estimate fuzzy regression parameters without using any XXXXX or heavy fuzzy arithmetic operations', ['error measures', 'mathematical programming'])\n", "('NONE', 'the tsk-type XXXXX uses flann in the consequent part of the XXXXX for improved mapping', ['fuzzy inference system', 'fuzzy rules'])\n", "('NONE', 'the XXXXX , asymmetric shock by XXXXX of XXXXX are important for forecasting', ['leverage effect', 'egarch model'])\n", "('NONE', 'the XXXXX , asymmetric shock by XXXXX of XXXXX are important for forecasting', ['leverage effect', 'egarch model'])\n", "('NONE', 'a XXXXX based XXXXX is used for egarch-flann model parameters', ['differential evolution', 'learning strategy'])\n", "('method used for task', 'a XXXXX based learning strategy is used for egarch-flann XXXXX', ['differential evolution', 'model parameters'])\n", "('method used for task', 'a differential evolution based XXXXX is used for egarch-flann XXXXX', ['learning strategy', 'model parameters'])\n", "('NONE', 'the proposed method outperformed the conventional methods ( the multi-layer autoencoder , the denoising autoencoder , and kernel pca ) for the XXXXX on XXXXX', ['synthetic data', 'linear regression task'])\n", "('NONE', 'in the XXXXX on ion conductivity data of bulk materials and hydrogen storage materials , the good XXXXX was obtained in terms of the latent data uniformed by the proposed method', ['linear regression task', 'fitting performance'])\n", "('NONE', 'proposed methodology uses the current amplitudes signal in XXXXX as the inputs of the XXXXX for fault diagnosis', ['time domain', 'multi-agent system'])\n", "('NONE', 'proposed methodology uses the current amplitudes signal in XXXXX as the inputs of the multi-agent system for XXXXX', ['time domain', 'fault diagnosis'])\n", "('method used for task', 'proposed methodology uses the current amplitudes signal in time domain as the inputs of the XXXXX for XXXXX', ['multi-agent system', 'fault diagnosis'])\n", "('NONE', 'the XXXXX incorporates XXXXX with better results for each type of fault', ['multi-agent system', 'pattern recognition techniques'])\n", "('NONE', 'experimental results gathered from three-phase XXXXX operating with different XXXXX and fed under unbalance voltage are provided', ['induction motors', 'load conditions'])\n", "('method used for task', 'we propose a XXXXX based on the XXXXX and semivalue', ['betweenness centrality', 'shapley value'])\n", "('NONE', 'we provide an XXXXX of algorithms on real-life and XXXXX', ['empirical evaluation', 'random graphs'])\n", "('NONE', 'we present an architecture based on XXXXX that allows to build a portable classifier with XXXXX , and we reach state-of-the-art performance', ['feature selection', 'minimum cost'])\n", "('NONE', 'XXXXX exploits XXXXX and all available mined information', ['domain knowledge', 'process comparison'])\n", "('NONE', 'tests in XXXXX show that , with respect to previously published metrics our approach generates outputs closer to those of a XXXXX expert ; the framework can support experts in answering key XXXXX', ['stroke management', 'research questions'])\n", "('NONE', 'tests in XXXXX show that , with respect to previously published metrics our approach generates outputs closer to those of a XXXXX expert ; the framework can support experts in answering key XXXXX', ['stroke management', 'research questions'])\n", "('NONE', 'XXXXX of common entries and similar XXXXX in two different pathways', ['automatic identification', 'reaction paths'])\n", "('method used for task', 'the proposed methodology , visual diagnostic dss , employs ml algorithms and XXXXX for XXXXX in medical genetics', ['automated diagnosis', 'image processing techniques'])\n", "('method used for task', 'nicesim provides a XXXXX for simulation after XXXXX', ['graphical user interface', 'model building'])\n", "('method used for task', 'we focus on short-term XXXXX considering complete surgery flow and the XXXXX involved', ['multiple resources', 'surgery scheduling problem'])\n", "('NONE', 'we propose a modified XXXXX with a two-level ant XXXXX and specific aco mechanism', ['graph model', 'aco algorithm'])\n", "('NONE', 'we proposed improved XXXXX measures of simplified neutrosophic sets ( snss ) based on cosine function , including single valued neutrosophic XXXXX measures and interval neutrosophic XXXXX measures , to overcome some disadvantages of existing XXXXX measures of snss', ['cosine similarity', 'cosine similarity measures'])\n", "('NONE', 'we proposed improved XXXXX measures of simplified neutrosophic sets ( snss ) based on cosine function , including single valued neutrosophic XXXXX measures and interval neutrosophic XXXXX measures , to overcome some disadvantages of existing XXXXX measures of snss', ['cosine similarity', 'cosine similarity measures'])\n", "('NONE', 'we proposed improved XXXXX measures of simplified neutrosophic sets ( snss ) based on cosine function , including single valued neutrosophic XXXXX measures and interval neutrosophic XXXXX measures , to overcome some disadvantages of existing XXXXX measures of snss', ['cosine similarity', 'cosine similarity measures'])\n", "('NONE', 'we proposed improved XXXXX measures of simplified neutrosophic sets ( snss ) based on cosine function , including single valued neutrosophic XXXXX measures and interval neutrosophic XXXXX measures , to overcome some disadvantages of existing XXXXX measures of snss', ['cosine similarity', 'cosine similarity measures'])\n", "('NONE', 'we proposed improved XXXXX measures of simplified neutrosophic sets ( snss ) based on cosine function , including single valued neutrosophic XXXXX measures and interval neutrosophic XXXXX measures , to overcome some disadvantages of existing XXXXX measures of snss', ['cosine similarity', 'cosine similarity measures'])\n", "('NONE', 'we proposed improved XXXXX measures of simplified neutrosophic sets ( snss ) based on cosine function , including single valued neutrosophic XXXXX measures and interval neutrosophic XXXXX measures , to overcome some disadvantages of existing XXXXX measures of snss', ['cosine similarity', 'cosine similarity measures'])\n", "('NONE', 'we proposed improved XXXXX measures of simplified neutrosophic sets ( snss ) based on cosine function , including single valued neutrosophic XXXXX measures and interval neutrosophic XXXXX measures , to overcome some disadvantages of existing XXXXX measures of snss', ['cosine similarity', 'cosine similarity measures'])\n", "('NONE', 'we proposed improved XXXXX measures of simplified neutrosophic sets ( snss ) based on cosine function , including single valued neutrosophic XXXXX measures and interval neutrosophic XXXXX measures , to overcome some disadvantages of existing XXXXX measures of snss', ['cosine similarity', 'cosine similarity measures'])\n", "('NONE', 'we proposed improved XXXXX measures of simplified neutrosophic sets ( snss ) based on cosine function , including single valued neutrosophic XXXXX measures and interval neutrosophic XXXXX measures , to overcome some disadvantages of existing XXXXX measures of snss', ['cosine similarity', 'cosine similarity measures'])\n", "('NONE', 'we proposed improved XXXXX measures of simplified neutrosophic sets ( snss ) based on cosine function , including single valued neutrosophic XXXXX measures and interval neutrosophic XXXXX measures , to overcome some disadvantages of existing XXXXX measures of snss', ['cosine similarity', 'cosine similarity measures'])\n", "('NONE', 'we proposed improved XXXXX measures of simplified neutrosophic sets ( snss ) based on cosine function , including single valued neutrosophic XXXXX measures and interval neutrosophic XXXXX measures , to overcome some disadvantages of existing XXXXX measures of snss', ['cosine similarity', 'cosine similarity measures'])\n", "('NONE', 'we proposed improved XXXXX measures of simplified neutrosophic sets ( snss ) based on cosine function , including single valued neutrosophic XXXXX measures and interval neutrosophic XXXXX measures , to overcome some disadvantages of existing XXXXX measures of snss', ['cosine similarity', 'cosine similarity measures'])\n", "('method used for task', 'we presented a XXXXX method based on the improved cosine similarity measures to solve XXXXX problems with simplified neutrosophic information', ['medical diagnosis', 'medical diagnosis method'])\n", "('method used for task', 'we presented a XXXXX method based on the improved XXXXX to solve XXXXX problems with simplified neutrosophic information', ['medical diagnosis', 'cosine similarity measures'])\n", "('method used for task', 'we presented a XXXXX method based on the improved cosine similarity measures to solve XXXXX problems with simplified neutrosophic information', ['medical diagnosis', 'medical diagnosis problems'])\n", "('method used for task', 'we presented a XXXXX based on the improved XXXXX to solve medical diagnosis problems with simplified neutrosophic information', ['medical diagnosis method', 'cosine similarity measures'])\n", "('method used for task', 'we presented a XXXXX based on the improved cosine similarity measures to solve XXXXX with simplified neutrosophic information', ['medical diagnosis method', 'medical diagnosis problems'])\n", "('method used for task', 'we presented a medical diagnosis method based on the improved XXXXX to solve XXXXX with simplified neutrosophic information', ['cosine similarity measures', 'medical diagnosis problems'])\n", "('method used for task', 'two XXXXX were given to show the effectiveness and rationality of the XXXXX using the improved cosine similarity measures', ['medical diagnosis problems', 'diagnosis method'])\n", "('NONE', 'two XXXXX were given to show the effectiveness and rationality of the diagnosis method using the improved XXXXX', ['medical diagnosis problems', 'cosine similarity measures'])\n", "('NONE', 'two medical diagnosis problems were given to show the effectiveness and rationality of the XXXXX using the improved XXXXX', ['diagnosis method', 'cosine similarity measures'])\n", "('method used for task', 'XXXXX ontology is a tool to create XXXXX of hci assessment', ['knowledge bases', 'investigation model'])\n", "('NONE', 'a semi-online XXXXX in the XXXXX is studied', ['patient scheduling problem', 'pathology laboratory'])\n", "('NONE', 'the approach reduces XXXXX of patients and improves XXXXX', ['waiting time', 'operations efficiency'])\n", "('method used for task', 'XXXXX approaches are evaluated for assigning XXXXX to emrs', ['supervised learning', 'diagnosis codes'])\n", "('NONE', 'we propose an approach to XXXXX in XXXXX based on grey relational analysis and d-s theory of evidence', ['fuzzy soft set', 'decision making'])\n", "('method used for task', 'we develop a XXXXX to integrate XXXXX with the basic rs model', ['hybrid model', 'word similarity'])\n", "('NONE', 'we improved XXXXX recall using suffix patterns evolved by XXXXX', ['genetic programming', 'drug ner'])\n", "('method used for task', 'prior XXXXX is incorporated into the XXXXX', ['domain knowledge', 'selection procedure'])\n", "('method used for task', 'knowledge engeering for medical expert systems dominated the first decade of aime , while XXXXX and XXXXX prevailed thereafter', ['machine learning', 'data mining'])\n", "('NONE', 'promising directions for XXXXX are XXXXX , personalized medicine , evidence based medicine , business process modeling , process mining and nlp in social media', ['future research', 'big data'])\n", "('NONE', 'promising directions for XXXXX are big data , XXXXX , evidence based medicine , business process modeling , process mining and nlp in social media', ['future research', 'personalized medicine'])\n", "('NONE', 'promising directions for XXXXX are big data , personalized medicine , evidence based medicine , business process modeling , XXXXX and nlp in social media', ['future research', 'process mining'])\n", "('NONE', 'promising directions for XXXXX are big data , personalized medicine , evidence based medicine , business process modeling , process mining and nlp in XXXXX', ['future research', 'social media'])\n", "('NONE', 'promising directions for future research are XXXXX , XXXXX , evidence based medicine , business process modeling , process mining and nlp in social media', ['big data', 'personalized medicine'])\n", "('NONE', 'promising directions for future research are XXXXX , personalized medicine , evidence based medicine , business process modeling , XXXXX and nlp in social media', ['big data', 'process mining'])\n", "('NONE', 'promising directions for future research are XXXXX , personalized medicine , evidence based medicine , business process modeling , process mining and nlp in XXXXX', ['big data', 'social media'])\n", "('NONE', 'promising directions for future research are big data , XXXXX , evidence based medicine , business process modeling , XXXXX and nlp in social media', ['personalized medicine', 'process mining'])\n", "('NONE', 'promising directions for future research are big data , XXXXX , evidence based medicine , business process modeling , process mining and nlp in XXXXX', ['personalized medicine', 'social media'])\n", "('NONE', 'promising directions for future research are big data , personalized medicine , evidence based medicine , business process modeling , XXXXX and nlp in XXXXX', ['process mining', 'social media'])\n", "('method used for task', 'XXXXX for simultaneous XXXXX and classifier weighting', ['evolutionary algorithm', 'feature selection'])\n", "('NONE', 'we encountered a clinical demand to process XXXXX within XXXXX medical logic modules', ['complex data', 'arden syntax'])\n", "('NONE', 'we encountered a clinical demand to process XXXXX within arden syntax medical XXXXX', ['complex data', 'logic modules'])\n", "('NONE', 'we encountered a clinical demand to process complex data within XXXXX medical XXXXX', ['arden syntax', 'logic modules'])\n", "('NONE', 'in 2004 we developed a pediatric computer based XXXXX , chica , using XXXXX medical logic modules ( mlm )', ['clinical decision support system', 'arden syntax'])\n", "('NONE', 'in 2004 we developed a pediatric computer based XXXXX , chica , using arden syntax medical XXXXX ( mlm )', ['clinical decision support system', 'logic modules'])\n", "('NONE', 'in 2004 we developed a pediatric computer based clinical decision support system , chica , using XXXXX medical XXXXX ( mlm )', ['arden syntax', 'logic modules'])\n", "('NONE', 'such a system can be used to combine data generated by XXXXX apps with XXXXX', ['clinical data', 'home monitoring'])\n", "('NONE', 'we examine its importance by an XXXXX of four XXXXX', ['empirical study', 'feature selection methods'])\n", "('method used for task', 'we use a XXXXX to improve XXXXX based on timing information', ['probabilistic model', 'classification accuracy'])\n", "('method used for task', 'the paper studies several XXXXX ( nlp ) techniques to extract predictors from uncoded data in XXXXX ( emrs )', ['natural language processing', 'electronic medical records'])\n", "('NONE', 'the XXXXX several XXXXX ( nlp ) techniques to extract predictors from uncoded data in electronic medical records ( emrs )', ['natural language processing', 'paper studies'])\n", "('method used for task', 'the XXXXX several natural language processing ( nlp ) techniques to extract predictors from uncoded data in XXXXX ( emrs )', ['electronic medical records', 'paper studies'])\n", "('NONE', 'the results show that some of the XXXXX studied can complement the coded emr data , and hence , result in improved XXXXX', ['predictive models', 'nlp techniques'])\n", "('method used for task', 'the objective is to provide real-world XXXXX and XXXXX', ['decision support', 'risk management'])\n", "('method used for task', 'both applications provide improvements in XXXXX and XXXXX', ['predictive accuracy', 'decision support'])\n", "('NONE', 'the developed XXXXX generates the parameters of a XXXXX by capturing the subjective and intuitive knowledge of medical physicists', ['cbr system', 'treatment plan'])\n", "('NONE', 'in this research we focus on the adaptation stage of the XXXXX in which the solution ( XXXXX ) of the retrieved case is adapted to meet the needs of the new case ( patient ) by considering differences between the retrieved and new cases', ['cbr system', 'treatment plan'])\n", "('NONE', 'the XXXXX was deployed alongside the XXXXX to evaluate its ability to discover rules that complement the existing rule set', ['learning module', 'baseline system'])\n", "('NONE', 'the XXXXX extracted clinically relevant rules , some of which extend the XXXXX of the baseline system', ['knowledge base', 'learning module'])\n", "('NONE', 'the learning module extracted clinically relevant rules , some of which extend the XXXXX of the XXXXX', ['knowledge base', 'baseline system'])\n", "('NONE', 'the XXXXX extracted clinically relevant rules , some of which extend the knowledge base of the XXXXX', ['learning module', 'baseline system'])\n", "('NONE', 'used XXXXX to determine multiple level of context in free living and contextualize XXXXX', ['machine learning methods', 'heart rate data'])\n", "('NONE', 'we develop a XXXXX with the XXXXX', ['particle filter', 'dirichlet distribution'])\n", "('NONE', 'XXXXX can be optimized by carefully XXXXX the workload', ['parallel performance', 'load balancing'])\n", "('method used for task', 'XXXXX is used for the XXXXX', ['comparative study', 'real time data'])\n", "('NONE', 'it integrates vo XXXXX as first-class XXXXX in taverna scientific workflows', ['web services', 'building blocks'])\n", "('NONE', 'the high XXXXX can be got based on XXXXX', ['clinical data', 'prediction performance'])\n", "('method used for task', 'a touching nuclei XXXXX based on XXXXX and concave vertex graph is proposed to quantify the total number of cancer nuclei in each class', ['watershed algorithm', 'separation method'])\n", "('NONE', 'XXXXX learn from XXXXX and multibody model solutions', ['neural networks', 'finite element'])\n", "('NONE', 'developed a novel framework to apply sparse and overcomplete representations for the removal of XXXXX in real XXXXX', ['speckle noise', 'ultrasound images'])\n", "('NONE', 'we proposed a combinational feature extraction approach based on some XXXXX extracted from electro cardio graph ( ecg ) XXXXX ( rps ) and usually used frequency domain features for detection of sleep apnoea', ['nonlinear features', 'reconstructed phase space'])\n", "('method used for task', 'here 6 XXXXX extracted from ecg rps are combined with 3 frequency based features to reconstruct final XXXXX', ['nonlinear features', 'feature set'])\n", "('NONE', 'the XXXXX consist of detrended fluctuation analysis ( dfa ) , correlation dimensions ( cd ) , 3 large XXXXX ( lles ) and spectral entropy ( se )', ['nonlinear features', 'lyapunov exponents'])\n", "('NONE', 'the XXXXX consist of detrended XXXXX ( dfa ) , correlation dimensions ( cd ) , 3 large lyapunov exponents ( lles ) and spectral entropy ( se )', ['nonlinear features', 'fluctuation analysis'])\n", "('NONE', 'the XXXXX consist of detrended fluctuation analysis ( dfa ) , XXXXX ( cd ) , 3 large lyapunov exponents ( lles ) and spectral entropy ( se )', ['nonlinear features', 'correlation dimensions'])\n", "('NONE', 'the nonlinear features consist of detrended XXXXX ( dfa ) , correlation dimensions ( cd ) , 3 large XXXXX ( lles ) and spectral entropy ( se )', ['lyapunov exponents', 'fluctuation analysis'])\n", "('NONE', 'the nonlinear features consist of detrended fluctuation analysis ( dfa ) , XXXXX ( cd ) , 3 large XXXXX ( lles ) and spectral entropy ( se )', ['lyapunov exponents', 'correlation dimensions'])\n", "('NONE', 'the nonlinear features consist of detrended XXXXX ( dfa ) , XXXXX ( cd ) , 3 large lyapunov exponents ( lles ) and spectral entropy ( se )', ['fluctuation analysis', 'correlation dimensions'])\n", "('NONE', 'a XXXXX ( fmm ) based XXXXX is proposed to interpolate the missing voxels', ['fast marching method', 'reconstruction algorithm'])\n", "('method used for task', 'the fmm XXXXX could preserve relatively sharper edges and more XXXXX in the reconstructed image slices', ['reconstruction algorithm', 'texture patterns'])\n", "('method used for task', 'the fmm XXXXX runs efficiently and requires less XXXXX', ['reconstruction algorithm', 'computation time'])\n", "('NONE', 'we classify different sleep stages with 41 XXXXX through XXXXX', ['random forest', 'hrv features'])\n", "('method used for task', 'we propose to extract texture feature descriptors based on XXXXX to characterize breast tumors in XXXXX', ['shearlet transform', 'ultrasound images'])\n", "('method used for task', 'novel XXXXX based on XXXXX and nonlinear transformation', ['feature extraction', 'tensor decomposition'])\n", "('NONE', 'the method can escape from XXXXX by using a fairly large XXXXX during freely deforming', ['local minima', 'detection range'])\n", "('method used for task', 'investigated XXXXX and model estimation methods for mr XXXXX', ['statistical models', 'image segmentation'])\n", "('method used for task', 'investigated XXXXX and XXXXX methods for mr image segmentation', ['statistical models', 'model estimation'])\n", "('method used for task', 'investigated statistical models and XXXXX methods for mr XXXXX', ['image segmentation', 'model estimation'])\n", "('NONE', 'XXXXX of eeg signal were extracted by XXXXX ( csp )', ['feature vectors', 'common spatial patterns'])\n", "('NONE', 'XXXXX were classified by XXXXX ( svm )', ['feature vectors', 'support vector machine'])\n", "('method used for task', 'we propose a multi-step directional ggvf XXXXX for XXXXX', ['snake model', 'target tumor segmentation'])\n", "('method used for task', 'for correlated rhythms , an XXXXX exists between bandwidth and XXXXX', ['spatial resolution', 'inverse relationship'])\n", "('method used for task', 'help the researchers to develop a new XXXXX for XXXXX', ['mr images', 'denoising method'])\n", "('method used for task', 'helps to choose the best XXXXX for further XXXXX', ['denoising method', 'image processing methods'])\n", "('NONE', 'XXXXX are XXXXX', ['respiratory sounds', 'chaotic signals'])\n", "('method used for task', 'a new XXXXX method is proposed for XXXXX in XXXXX', ['noise reduction', 'respiratory sounds'])\n", "('method used for task', 'a new XXXXX method is proposed for XXXXX in respiratory sounds', ['noise reduction', 'noise reduction method'])\n", "('method used for task', 'a new XXXXX is proposed for noise reduction in XXXXX', ['respiratory sounds', 'noise reduction method'])\n", "('NONE', 'the proposed method is based on new nonlinear XXXXX as controller that their parameter tunes by XXXXX', ['convex function', 'genetic algorithm'])\n", "('method used for task', 'a framework on wavelet-based XXXXX and extreme XXXXX is proposed for the seizure detection', ['nonlinear features', 'machine learning'])\n", "('method used for task', 'a framework on wavelet-based XXXXX and extreme machine learning is proposed for the XXXXX', ['nonlinear features', 'seizure detection'])\n", "('method used for task', 'a framework on wavelet-based nonlinear features and extreme XXXXX is proposed for the XXXXX', ['machine learning', 'seizure detection'])\n", "('method used for task', 'applications in the natural XXXXX and pathological XXXXX', ['speech synthesis', 'speech recognition'])\n", "('NONE', 'the paper presents a new method for XXXXX in ultrasound XXXXX', ['speckle reduction', 'medical images'])\n", "('method used for task', 'the proposed method combines the XXXXX and ripplet thresholding to remove the speckles in XXXXX', ['bilateral filter', 'ultrasound images'])\n", "('method used for task', 'this improved XXXXX is used as an input to evolve the initial XXXXX', ['edge information', 'level set function'])\n", "('NONE', 'we extend the method to afford XXXXX of new models in XXXXX', ['unsupervised learning', 'real time'])\n", "('NONE', \"a XXXXX estimated the time varying XXXXX representing the windkessel model 's output signal\", ['kalman filter', 'fourier series'])\n", "('method used for task', 'XXXXX reconstruct method is used to reconstruct the XXXXX of XXXXX', ['phase space', 'eeg signals'])\n", "('method used for task', 'XXXXX reconstruct method is used to reconstruct the XXXXX of XXXXX', ['phase space', 'eeg signals'])\n", "('NONE', 'the features are extracted by amplitude–frequency XXXXX in the XXXXX', ['phase space', 'analysis method'])\n", "('method used for task', 'using an fpga platform , we showed that the cs-based XXXXX , compared to a low-power wavelet-based XXXXX , can largely reduce XXXXX and save other computing resources', ['power consumption', 'compression method'])\n", "('method used for task', 'using an fpga platform , we showed that the cs-based XXXXX , compared to a low-power wavelet-based XXXXX , can largely reduce XXXXX and save other computing resources', ['power consumption', 'compression method'])\n", "('method used for task', 'design of an energy optimal trajectory planner , based on XXXXX to decide the optimal path for the XXXXX to move towards the target', ['differential evolution', 'robot arm'])\n", "('NONE', 'the XXXXX of the simulated XXXXX reaching the target is 85 %', ['robot arm', 'success rate'])\n", "('method used for task', 'we proposed a novel XXXXX rejection approach to suppress the XXXXX signals induced by the slowly moving vessel walls and tissues in composite XXXXX', ['doppler ultrasound signals', 'quadrature clutter'])\n", "('method used for task', 'we proposed a novel XXXXX rejection approach to suppress the XXXXX signals induced by the slowly moving vessel walls and tissues in composite XXXXX', ['doppler ultrasound signals', 'quadrature clutter'])\n", "('method used for task', 'the proposed XXXXX approach is based on multivariate empirical XXXXX ( memd )', ['clutter rejection', 'mode decomposition'])\n", "('NONE', 'the XXXXX between the quadrature demodulated XXXXX is keeping well during the decomposition by memd', ['phase information', 'doppler ultrasound signals'])\n", "('NONE', 'we quantify contrast agent XXXXX in plaque using XXXXX', ['spatial distribution', 'texture features'])\n", "('NONE', 'the XXXXX of histogram are extracted for XXXXX', ['statistical features', 'svm classification'])\n", "('NONE', 'a new angiogram background suppression method based on 3d XXXXX is proposed in the context of sparsity regularized XXXXX of coronary artery', ['vessel segmentation', 'iterative reconstruction'])\n", "('NONE', 'a new angiogram background suppression method based on 3d XXXXX is proposed in the context of sparsity regularized iterative reconstruction of XXXXX', ['vessel segmentation', 'coronary artery'])\n", "('NONE', 'a new angiogram background suppression method based on 3d vessel segmentation is proposed in the context of sparsity regularized XXXXX of XXXXX', ['iterative reconstruction', 'coronary artery'])\n", "('NONE', 'several experiments are performed to quantitatively evaluate the proposed method and experimental results show that it can effectively improve the XXXXX of XXXXX', ['coronary artery', 'reconstruction quality'])\n", "('NONE', 'XXXXX of XXXXX diabetic göttingen minipigs', ['model development', 'glucose metabolism'])\n", "('method used for task', 'development of XXXXX for future XXXXX', ['mathematical model', 'control design'])\n", "('NONE', 'XXXXX , especially the dynamic change of arm position , have adverse effect on surface electromyography pattern recognition of wrist or hand motions , which results in very high XXXXX', ['arm movements', 'classification errors'])\n", "('method used for task', 'we adopted three metrics to quantify the changes of characteristics in the XXXXX due to XXXXX', ['feature space', 'arm movements'])\n", "('method used for task', 'a robust training scheme was used to eliminate this effect and a novel classifier , conditional XXXXX , was proposed to improve the XXXXX under this training scheme', ['gaussian mixture model', 'classification performance'])\n", "('method used for task', 'a robust XXXXX was used to eliminate this effect and a novel classifier , conditional XXXXX , was proposed to improve the classification performance under this XXXXX', ['gaussian mixture model', 'training scheme'])\n", "('method used for task', 'a robust XXXXX was used to eliminate this effect and a novel classifier , conditional XXXXX , was proposed to improve the classification performance under this XXXXX', ['gaussian mixture model', 'training scheme'])\n", "('method used for task', 'a robust XXXXX was used to eliminate this effect and a novel classifier , conditional gaussian mixture model , was proposed to improve the XXXXX under this XXXXX', ['classification performance', 'training scheme'])\n", "('method used for task', 'a robust XXXXX was used to eliminate this effect and a novel classifier , conditional gaussian mixture model , was proposed to improve the XXXXX under this XXXXX', ['classification performance', 'training scheme'])\n", "('method used for task', 'boundaries of XXXXX ( s1 , s2 , s3 , s4 ) and murmurs are automatically determined using instantaneous XXXXX of the envelope', ['heart sounds', 'phase waveform'])\n", "('NONE', 'XXXXX from a contour integration task are analyzed with empirical XXXXX', ['fmri data', 'mode decomposition'])\n", "('NONE', 'introduction of a switching control method and XXXXX by means of two XXXXX of the porcine glucose metabolism', ['controller design', 'mathematical models'])\n", "('NONE', 'introduction of a switching control method and XXXXX by means of two mathematical models of the porcine XXXXX', ['controller design', 'glucose metabolism'])\n", "('method used for task', 'introduction of a XXXXX and XXXXX by means of two mathematical models of the porcine glucose metabolism', ['controller design', 'switching control method'])\n", "('NONE', 'introduction of a switching control method and controller design by means of two XXXXX of the porcine XXXXX', ['mathematical models', 'glucose metabolism'])\n", "('NONE', 'introduction of a XXXXX and controller design by means of two XXXXX of the porcine glucose metabolism', ['mathematical models', 'switching control method'])\n", "('NONE', 'introduction of a XXXXX and controller design by means of two mathematical models of the porcine XXXXX', ['glucose metabolism', 'switching control method'])\n", "('NONE', 'XXXXX of XXXXX by animal trials taking into account control performance and technical limitations', ['experimental evaluation', 'control method'])\n", "('NONE', 'XXXXX of control method by animal trials taking into account XXXXX and technical limitations', ['experimental evaluation', 'control performance'])\n", "('NONE', 'experimental evaluation of XXXXX by animal trials taking into account XXXXX and technical limitations', ['control method', 'control performance'])\n", "('method used for task', 'estimated parameters of the ar-garch model formed the new XXXXX for the XXXXX', ['feature set', 'pattern classification'])\n", "('NONE', 'the possibility to reconstruct a XXXXX from XXXXX in order to characterize it from a dynamical point of view is performed', ['mathematical model', 'experimental data'])\n", "('method used for task', 'the XXXXX , the fractal and XXXXX could be used for classification feature', ['time lag', 'correlation dimensions'])\n", "('NONE', 'the significance of the measurements is assessment based on the XXXXX , the total XXXXX ( the level of noise plus non-linearities ) , the experimental errors and the effect of using different sensors', ['standard deviation', 'noise level'])\n", "('NONE', 'the paper presents an algorithm for XXXXX in XXXXX', ['speckle reduction', 'ultrasound images'])\n", "('NONE', 'the proposed method is based on non-linear adaptive XXXXX in nonsubsampled XXXXX', ['diffusion model', 'shearlet domain'])\n", "('method used for task', 'the proposed method provides better XXXXX and edge preservation performance for ultrasound XXXXX as compared to others', ['noise suppression', 'medical images'])\n", "('NONE', 'we perform an XXXXX of XXXXX', ['automatic analysis', 'speech signals'])\n", "('NONE', 'XXXXX exhibit longer pauses between each XXXXX', ['pd patients', 'sentence repetition'])\n", "('method used for task', 'redundant information reduced significantly which saves XXXXX and improves XXXXX', ['computational time', 'image quality'])\n", "('NONE', 'we use a set of three XXXXX as a dynamical model of XXXXX', ['differential equations', 'ecg signals'])\n", "('method used for task', 'two classifiers are constructed to classify XXXXX according to the XXXXX', ['ecg signals', 'parameter changes'])\n", "('method used for task', 'the method uses XXXXX and XXXXX for the extraction of oriented patterns and linear structures representing breast tissues', ['gabor filters', 'radon transform'])\n", "('method used for task', 'XXXXX for the XXXXX based on sparse representation', ['compression algorithm', 'ecg signals'])\n", "('method used for task', 'XXXXX for the ecg signals based on XXXXX', ['compression algorithm', 'sparse representation'])\n", "('method used for task', 'compression algorithm for the XXXXX based on XXXXX', ['ecg signals', 'sparse representation'])\n", "('NONE', 'XXXXX are better in terms of power spectra and XXXXX', ['stochastic models', 'information entropy'])\n", "('NONE', 'XXXXX are better in terms of XXXXX and information entropy', ['stochastic models', 'power spectra'])\n", "('method used for task', 'stochastic models are better in terms of XXXXX and XXXXX', ['information entropy', 'power spectra'])\n", "('method used for task', 'bhattacharyya space algorithm , t-test , wilcoxon test , receiver operating curve ( roc ) , and XXXXX are used for XXXXX', ['feature ranking', 'entropy method'])\n", "('method used for task', 'the comparation is given between the XXXXX strategy and constant XXXXX , which shows the XXXXX strategy is better than constant XXXXX', ['optimal control', 'control strategy'])\n", "('method used for task', 'the comparation is given between the XXXXX strategy and constant control strategy , which shows the XXXXX strategy is better than constant control strategy', ['optimal control', 'optimal control strategy'])\n", "('method used for task', 'the comparation is given between the XXXXX strategy and constant control strategy , which shows the XXXXX strategy is better than constant control strategy', ['optimal control', 'optimal control strategy'])\n", "('method used for task', 'the comparation is given between the optimal XXXXX and constant XXXXX , which shows the optimal XXXXX is better than constant XXXXX', ['control strategy', 'optimal control strategy'])\n", "('method used for task', 'the comparation is given between the optimal XXXXX and constant XXXXX , which shows the optimal XXXXX is better than constant XXXXX', ['control strategy', 'optimal control strategy'])\n", "('NONE', 'increasing XXXXX parameter will decrease the XXXXX', ['model performance', 'constraint window'])\n", "('NONE', 'stochastic model predictive ( stomp ) is a glycaemic XXXXX that combines the probabilistic , stochastic XXXXX of previous methods ( star ) with model predictive control , for ease of tuning', ['control protocol', 'forecasting methods'])\n", "('method used for task', 'this paper presents a hybrid XXXXX for ultrasound XXXXX', ['medical images', 'segmentation method'])\n", "('NONE', 'the proposed algorithm can distinguish among six classes of XXXXX using two temporal eeg sensors with XXXXX and low-latency using a single-trial to make a decision', ['eye movements', 'high accuracy'])\n", "('NONE', 'XXXXX ( acms ) detecting XXXXX ( od ) boundaries are investigated', ['active contour models', 'optic disc'])\n", "('method used for task', 'images are subject to adaptive XXXXX and XXXXX', ['histogram equalization', 'morphological operations'])\n", "('NONE', 'we achieve better performance on simulated and XXXXX medical XXXXX', ['in vivo', 'ultrasound images'])\n", "('method used for task', 'we develop ground reaction force ( grf ) based XXXXX for XXXXX', ['human gait', 'asymmetry measure'])\n", "('method used for task', 'the proposed method is based on the XXXXX measure and the pcnn model for fusing the image coefficients in nonsubsampled XXXXX', ['activity level', 'shearlet domain'])\n", "('method used for task', 'cleaned real XXXXX are used for pattern recognition-based XXXXX', ['gesture recognition', 'emg signals'])\n", "('method used for task', 'XXXXX is improved for real emg with added XXXXX', ['classification accuracy', 'white noise'])\n", "('NONE', 'a new algorithm named ibfsvm is proposed , which is based on fuzzy XXXXX and can be used in XXXXX', ['support vector', 'imbalanced classification'])\n", "('method used for task', 'the multi-ganglion ann based XXXXX ( annfl ) method is an unsupervised XXXXX', ['feature learning', 'feature extraction method'])\n", "('method used for task', 'a novel method based on XXXXX is proposed for XXXXX from eeg data , simultaneously recorded with fmri', ['dictionary learning', 'bcg removal'])\n", "('NONE', 'a new XXXXX , having the generic XXXXX as its core , is proposed', ['cost function', 'dictionary learning'])\n", "('method used for task', 'the hms based entropy and XXXXX are extracted for XXXXX', ['svm classification', 'energy features'])\n", "('method used for task', 'an optimal XXXXX is chosen by formal XXXXX', ['feature vector', 'statistical testing'])\n", "('method used for task', 'we developed a new XXXXX to detect XXXXX from ecg', ['statistical model', 'atrial fibrillation'])\n", "('NONE', 'an automatic detection algorithm for detection of pathologic voices is suggested . residual signals extracted by applying XXXXX on wavelet sub-bands provide a highly reliable vocal disorders detection scheme . by employing XXXXX and support vector machine , a great classification is achieved', ['linear prediction', 'linear discriminant analysis'])\n", "('method used for task', 'an automatic detection algorithm for detection of pathologic voices is suggested . residual signals extracted by applying XXXXX on wavelet sub-bands provide a highly reliable vocal disorders detection scheme . by employing linear discriminant analysis and XXXXX , a great classification is achieved', ['linear prediction', 'support vector machine'])\n", "('method used for task', 'an automatic detection algorithm for detection of pathologic voices is suggested . residual signals extracted by applying linear prediction on wavelet sub-bands provide a highly reliable vocal disorders detection scheme . by employing XXXXX and XXXXX , a great classification is achieved', ['linear discriminant analysis', 'support vector machine'])\n", "('NONE', 'motivation : synchronize XXXXX from several XXXXX is a difficult task', ['biomedical signals', 'acquisition systems'])\n", "('method used for task', 'methodology : use of XXXXX and correlation methods to synchronize XXXXX', ['white noise', 'biomedical signals'])\n", "('method used for task', 'XXXXX largely compresses neural signals and has a small XXXXX', ['reconstruction error', 'mdc matrix'])\n", "('NONE', 'XXXXX can be used over a large range of the XXXXX for compression', ['mdc matrix', 'sampling rate'])\n", "('method used for task', 'an improved XXXXX for multimodal XXXXX', ['retinal images', 'registration framework'])\n", "('method used for task', 'we propose two adaptive compression paradigms for the XXXXX ( dwt ) and XXXXX ( cs ) compression techniques', ['discrete wavelet transform', 'compressive sensing'])\n", "('method used for task', 'we propose two adaptive compression paradigms for the XXXXX ( dwt ) and compressive sensing ( cs ) XXXXX', ['discrete wavelet transform', 'compression techniques'])\n", "('NONE', 'we propose two adaptive compression paradigms for the discrete wavelet transform ( dwt ) and XXXXX ( cs ) XXXXX', ['compressive sensing', 'compression techniques'])\n", "('NONE', 'the paper presents a XXXXX of newborn cry signals based on XXXXX', ['hidden markov models', 'segmentation system'])\n", "('method used for task', 'the XXXXX and respiratory rate are estimated form the XXXXX envelope', ['heart rate', 'shannon energy'])\n", "('method used for task', 'a XXXXX for eeg XXXXX is proposed which combines permutation and lempel-ziv complexity ( plzc )', ['signal analysis', 'complexity measure'])\n", "('method used for task', 'combine XXXXX and wilcoxon statistics for eeg signal XXXXX', ['wavelet transform', 'feature extraction'])\n", "('method used for task', 'pulsations of XXXXX and XXXXX are correlated for 0.01–0.1hz', ['blood flow', 'skin temperature'])\n", "('NONE', 'finding best samples for lmmse estimation in XXXXX will improve the XXXXX', ['dct domain', 'denoising performance'])\n", "('method used for task', 'its performances were tested using both coupled XXXXX and clinical cardiovascular coupled signals , and were also compared with two existing cross entropy measures : cross XXXXX ( c-sampen ) and cross fuzzy entropy ( c-fuzzyen )', ['simulation models', 'sample entropy'])\n", "('method used for task', 'its performances were tested using both coupled XXXXX and clinical cardiovascular coupled signals , and were also compared with two existing cross entropy measures : cross sample entropy ( c-sampen ) and cross XXXXX ( c-fuzzyen )', ['simulation models', 'fuzzy entropy'])\n", "('method used for task', 'its performances were tested using both coupled simulation models and clinical cardiovascular coupled signals , and were also compared with two existing cross entropy measures : cross XXXXX ( c-sampen ) and cross XXXXX ( c-fuzzyen )', ['sample entropy', 'fuzzy entropy'])\n", "('NONE', 'XXXXX ( hr ) monitoring using wrist-type XXXXX during physical exercise was examined', ['heart rate', 'ppg signals'])\n", "('NONE', 'evaluation of XXXXX of XXXXX based classification ( src ) method', ['noise robustness', 'sparse representation'])\n", "('NONE', 'generation of new noisy XXXXX by adding two XXXXX into the eeg XXXXX', ['test signals', 'noise sources'])\n", "('NONE', 'generation of new noisy XXXXX by adding two XXXXX into the eeg XXXXX', ['noise sources', 'test signals'])\n", "('NONE', 'demonstration of the improved XXXXX of the src for noisy XXXXX and online dataset', ['classification accuracy', 'test signals'])\n", "('NONE', 'XXXXX increased with XXXXX', ['image quality', 'compressive sensing method'])\n", "('NONE', 'we use 3d-dwt to capture the 3d XXXXX of brain , use als-pca for XXXXX of dataset containing missing attributes', ['texture feature', 'feature reduction'])\n", "('NONE', 'an unsupervised , XXXXX , XXXXX for brain tumor segmentation is proposed', ['low cost', 'hybrid approach'])\n", "('method used for task', 'XXXXX is evaluated by standard viability assay while XXXXX required the development of a special algorithm', ['cell proliferation', 'cell morphology'])\n", "('NONE', 'our data show that ultrasounds induce cell differentiation affecting XXXXX as well as the ability of neurons and microglia to form XXXXX', ['complex networks', 'cell morphology'])\n", "('method used for task', 'we propose a multiscale svd based XXXXX for multilead XXXXX', ['compression algorithm', 'ecg data'])\n", "('method used for task', 'XXXXX and XXXXX are employed to improve performance of the system', ['genetic algorithm', 'linear discriminant analysis'])\n", "('NONE', 'XXXXX permit identifying characteristic points of the XXXXX', ['curvelet transform', 'ecg signal'])\n", "('NONE', 'detection of pvc in XXXXX using the fractional XXXXX outperforms detection using XXXXX', ['ecg signals', 'linear prediction'])\n", "('NONE', 'detection of pvc in XXXXX using the fractional XXXXX outperforms detection using XXXXX', ['ecg signals', 'linear prediction'])\n", "('method used for task', 'we propose an adaptive rv measure based fuzzy weighting XXXXX for XXXXX', ['subspace clustering', 'fmri data analysis'])\n", "('NONE', 'compared with the XXXXX , the normalized XXXXX can be used to locate the lesion in cortex , since the real cortex is far less-than-ideal model', ['gradient method', 'pca decomposition'])\n", "('NONE', 'a speckle reduction filter for XXXXX using XXXXX on coefficient of variation ( cv ) is proposed', ['ultrasound images', 'fuzzy logic'])\n", "('NONE', 'experiments based on well established database were carried out to evaluate the performance of the proposed approach . the results demonstrated better characterization of the movement assessment and motion recognition ability , with a XXXXX of 86.19 % , than the currently used methods such as XXXXX and pose normalization employed in motion recognition tasks', ['recognition rate', 'gaussian mixture models'])\n", "('NONE', 'the XXXXX is computationally efficient in the presence of high-resolution retinal XXXXX', ['fundus images', 'registration method'])\n", "('method used for task', 's-crc is tested for emg signal XXXXX for 10 XXXXX', ['pattern recognition', 'hand gestures'])\n", "('method used for task', 'in a former paper which is accepted in this journal ( which is under revision ) we implemented several filtering methods for hemodynamic XXXXX . there we showed in detail that the hemodynamic model is weakly non-linear . based on that finding we extended our researched by implementing a successful joint hemodynamic state and XXXXX . we also implemented the algorithm to a real fmri data set', ['state estimation', 'parameter estimation method'])\n", "('method used for task', 'in a former paper which is accepted in this journal ( which is under revision ) we implemented several filtering methods for hemodynamic XXXXX . there we showed in detail that the hemodynamic model is weakly non-linear . based on that finding we extended our researched by implementing a successful joint hemodynamic state and parameter estimation method . we also implemented the algorithm to a real XXXXX set', ['state estimation', 'fmri data'])\n", "('method used for task', 'in a former paper which is accepted in this journal ( which is under revision ) we implemented several filtering methods for hemodynamic state estimation . there we showed in detail that the hemodynamic model is weakly non-linear . based on that finding we extended our researched by implementing a successful joint hemodynamic state and XXXXX . we also implemented the algorithm to a real XXXXX set', ['parameter estimation method', 'fmri data'])\n", "('method used for task', 'we emphasized the following point in the literature . it is alleged that XXXXX is unrobust and poor in performance . we carefully examined this allegation under various state/measurement noise conditions . this allegation is not true for the widely used hemodynamic model in a wide range of process and measurement noise conditions . furthermore , several times XXXXX are alleged to outperform extended kalman filters . the alleged ekf is shown to be better in state estimation . we examined carefully and showed the contrary is true', ['extended kalman filter', 'particle filters'])\n", "('method used for task', 'we emphasized the following point in the literature . it is alleged that XXXXX is unrobust and poor in performance . we carefully examined this allegation under various state/measurement noise conditions . this allegation is not true for the widely used hemodynamic model in a wide range of process and measurement noise conditions . furthermore , several times particle filters are alleged to outperform extended kalman filters . the alleged ekf is shown to be better in XXXXX . we examined carefully and showed the contrary is true', ['extended kalman filter', 'state estimation'])\n", "('method used for task', 'we emphasized the following point in the literature . it is alleged that extended kalman filter is unrobust and poor in performance . we carefully examined this allegation under various state/measurement noise conditions . this allegation is not true for the widely used hemodynamic model in a wide range of process and measurement noise conditions . furthermore , several times XXXXX are alleged to outperform extended kalman filters . the alleged ekf is shown to be better in XXXXX . we examined carefully and showed the contrary is true', ['particle filters', 'state estimation'])\n", "('method used for task', 'the XXXXX is applied for the joint parameter and XXXXX . in the standard implementation of smoother there is no iteration . we compared iterative performance and noted substantial improvement in performance', ['iterative scheme', 'state estimation'])\n", "('NONE', '1/enhl negatively correlates with the XXXXX α of a de-trended XXXXX', ['scaling factor', 'fluctuation analysis'])\n", "('NONE', 'enhl and the XXXXX α reflect information about XXXXX of balance', ['scaling factor', 'motor control'])\n", "('method used for task', 'random forest , random tree , mlp network and XXXXX ( svm ) are employed for XXXXX', ['support vector machines', 'data classification'])\n", "('method used for task', 'random forest , XXXXX , mlp network and XXXXX ( svm ) are employed for data classification', ['support vector machines', 'random tree'])\n", "('method used for task', 'random forest , XXXXX , mlp network and support vector machines ( svm ) are employed for XXXXX', ['data classification', 'random tree'])\n", "('NONE', 'the XXXXX using the mit-bih arrhythmia database shows a high XXXXX of 97 %', ['performance evaluation', 'classification accuracy'])\n", "('NONE', 'the XXXXX using the mit-bih XXXXX shows a high classification accuracy of 97 %', ['performance evaluation', 'arrhythmia database'])\n", "('NONE', 'the performance evaluation using the mit-bih XXXXX shows a high XXXXX of 97 %', ['classification accuracy', 'arrhythmia database'])\n", "('NONE', 'the proposed work presents a novel framework for fast and fully XXXXX of od in XXXXX', ['automatic detection', 'fundus images'])\n", "('NONE', 'a new method for XXXXX of XXXXX is proposed', ['automatic segmentation', 'coronary arteries'])\n", "('method used for task', 'the high XXXXX achieved is represented by an area under XXXXX of az=0.961 with a training set of 40 images', ['detection rate', 'roc curve'])\n", "('NONE', 'an automatic multi-organ XXXXX of prostate XXXXX is proposed', ['magnetic resonance images', 'segmentation method'])\n", "('NONE', 'a 2d XXXXX is modeled using the inter-scale ratio of XXXXX', ['wavelet coefficients', 'gel image'])\n", "('NONE', 'the XXXXX obtained can help integrating a cgms into an XXXXX', ['artificial pancreas', 'error reduction'])\n", "('NONE', 'XXXXX of the XXXXX in frequency and time domain are derived', ['analytical solutions', 'dispersion model'])\n", "('NONE', 'XXXXX of the dispersion model in frequency and XXXXX are derived', ['analytical solutions', 'time domain'])\n", "('NONE', 'analytical solutions of the XXXXX in frequency and XXXXX are derived', ['dispersion model', 'time domain'])\n", "('method used for task', 'the work studies the XXXXX from wide band electrocardiogram ( ecg ) signals for classifying XXXXX ( mi ) stages', ['feature extraction', 'myocardial infarction'])\n", "('method used for task', 'we proposed a new efficient XXXXX for XXXXX', ['photoacoustic imaging', 'image reconstruction algorithm'])\n", "('NONE', 'the proposed XXXXX has better XXXXX to enter into long and thin indentations of medical images', ['snake model', 'convergence property'])\n", "('method used for task', 'the proposed XXXXX has better convergence property to enter into long and thin indentations of XXXXX', ['snake model', 'medical images'])\n", "('method used for task', 'the proposed snake model has better XXXXX to enter into long and thin indentations of XXXXX', ['convergence property', 'medical images'])\n", "('method used for task', 'warm XXXXX ( about 33°c ) should be maintained for reliable XXXXX', ['pulse oximetry', 'site temperature'])\n", "('method used for task', 'a compressive sampling pam system based on XXXXX ( cslrm-pam ) system has been set up to achieve fast XXXXX', ['data acquisition', 'low rank matrix completion'])\n", "('NONE', 'the final pam image is recovered directly from the compressive data by XXXXX without dramatic loss of XXXXX', ['low rank matrix completion', 'image resolution'])\n", "('NONE', 'XXXXX of XXXXX while running outdoors is deemed feasible', ['feedback control', 'heart rate'])\n", "('method used for task', 'on normal muscles , it performs better than other relevant algorithms , regarding XXXXX and quantitative XXXXX figures of merit', ['signal processing', 'muap waveform'])\n", "('method used for task', 'we review the XXXXX ( iqa ) for XXXXX', ['image quality assessment', 'medical images'])\n", "('NONE', 'majority of iqa was done on XXXXX ( mri ) , XXXXX ( ct ) , and ultrasonic images', ['magnetic resonance imaging', 'computed tomography'])\n", "('method used for task', 'the scheme was implemented in a XXXXX for XXXXX', ['software framework', 'mobile devices'])\n", "('NONE', 'the scale and shape parameters of the XXXXX are estimated using the XXXXX ( ml ) method', ['gamma distribution', 'maximum likelihood'])\n", "('NONE', 'a denoised estimate is made in the XXXXX by defining an XXXXX', ['input space', 'inverse mapping'])\n", "('NONE', 'XXXXX of 99.92 % and XXXXX of 99.73 %', ['classification accuracy', 'system reliability'])\n", "('NONE', 'XXXXX of XXXXX', ['search algorithm', 'circular code motifs'])\n", "('NONE', 'based on available XXXXX , the XXXXX is up to 90 %', ['experimental data', 'prediction accuracy'])\n", "('NONE', 'existing methods of pooling similar rare variants were based on quantities calculated from XXXXX . our method pools rare variants with similar XXXXX calculated based on the ratio of variant frequencies between cases and controls', ['relative risk', 'control group'])\n", "('NONE', 'we have used autodock vina to compare the XXXXX of potential inhibitors against an essential enzyme in XXXXX and its human homolog', ['plasmodium falciparum', 'binding affinities'])\n", "('method used for task', 'XXXXX optimizations need to be included in XXXXX', ['hydrogen bond network', 'simulation protocols'])\n", "('NONE', 'the role of XXXXX in complexes computed and biophysical property of XXXXX involved in complexes interface was considered', ['amino acids', 'water molecules'])\n", "('NONE', 'the XXXXX of XXXXX a was most significantly affected by the binding of the atp and peptide ligand', ['structure dynamics', 'protein kinase'])\n", "('NONE', 'XXXXX in the domains of the proteins show that repeats involved in the function of protein possess conserved XXXXX', ['sequence repeats', 'sequence motifs'])\n", "('method used for task', 'XXXXX programing is an effective method to evolve XXXXX ( odes ) model from observed time-series data', ['gene expression', 'ordinary differential equations'])\n", "('method used for task', 'XXXXX programing is an effective method to evolve ordinary differential equations ( odes ) model from observed XXXXX', ['gene expression', 'time-series data'])\n", "('NONE', 'gene expression programing is an effective method to evolve XXXXX ( odes ) model from observed XXXXX', ['ordinary differential equations', 'time-series data'])\n", "('method used for task', 'XXXXX for each vaccine modeled by XXXXX', ['tumor growth', 'artificial neural networks'])\n", "('NONE', 'we employed XXXXX , structural motif , XXXXX , and solvent accessibility calculation', ['secondary structure', 'conservation pattern'])\n", "('NONE', 'XXXXX revealed XXXXX of l-type pyruvate kinase n-domain in domains a and c of the protein', ['computational docking', 'binding sites'])\n", "('NONE', 'the algorithm basically utilizes XXXXX from XXXXX', ['phylogenetic tree', 'multiple sequence alignment'])\n", "('NONE', 'we propose a XXXXX for detection of XXXXX epistatic interaction in gwas', ['fast method', 'high order'])\n", "('NONE', 'a new XXXXX for chemicals using time-dependent cellular XXXXX is proposed', ['classification method', 'response profiles'])\n", "('NONE', 'a XXXXX machine-based XXXXX is proposed to predict conformational b-cell epitopes', ['support vector', 'ensemble method'])\n", "('method used for task', 'each gene pair was represented by incorporating the XXXXX and XXXXX', ['expression data', 'sequence information'])\n", "('method used for task', 'hierarchical closeness outperforms the other well-known structural XXXXX , particularly for cancer , hereditary , immune , and neurodegenerative disease-related genes in a human XXXXX', ['centrality measures', 'signaling network'])\n", "('NONE', 'establish our XXXXX basing on XXXXX', ['prediction model', 'support vector machine classifier'])\n", "('method used for task', 'we proposed a XXXXX to increase XXXXX', ['hybrid algorithm', 'metabolites production'])\n", "('NONE', 'we implement kinetic formalism of rene thomas along with XXXXX approaches to provide qualitative thresholds of entities in tlr3 associated biological regulatory network of XXXXX , satisfying the biological observations', ['model checking', 'dengue virus'])\n", "('NONE', 'different XXXXX of e296 ( d299 ) and d312 ( 315 ) in ompf and ompc promote the large-\\xadscale deviation in XXXXX and dynamics', ['protein structure', 'protonation states'])\n", "('method used for task', 'we proposed a XXXXX for protein–rna binding sites prediction by combining XXXXX and global features from protein sequence based on submodularity subset selection', ['computational method', 'local features'])\n", "('method used for task', 'we proposed a XXXXX for protein–rna binding sites prediction by combining local features and global features from XXXXX based on submodularity subset selection', ['computational method', 'protein sequence'])\n", "('method used for task', 'we proposed a computational method for protein–rna binding sites prediction by combining XXXXX and global features from XXXXX based on submodularity subset selection', ['local features', 'protein sequence'])\n", "('method used for task', 'proposed a novel basic XXXXX method used for ds XXXXX', ['evidence theory', 'probability assignment'])\n", "('NONE', 'we improved the XXXXX by adding a uniform XXXXX in the onlooker phase', ['abc algorithm', 'crossover operation'])\n", "('NONE', 'gnc is a XXXXX to the problem of XXXXX biological validation', ['robust solution', 'gene network'])\n", "('NONE', 'XXXXX of XXXXX were explored based on these degs', ['complex networks', 'hub genes'])\n", "('method used for task', 'XXXXX is accentuated by XXXXX', ['parkinson’s disease', 'insulin resistance'])\n", "('method used for task', 'we have incorporated these heuristics into a celebrated XXXXX called spici to get three new XXXXX', ['clustering algorithm', 'clustering algorithms'])\n", "('NONE', 'a method of identifying XXXXX based on moepga in XXXXX is proposed', ['protein complexes', 'ppi network'])\n", "('method used for task', 'here , XXXXX , dipeptide composition , correlation features , composition , transition and distribution and pseudo XXXXX are used to represent the XXXXX', ['amino acid composition', 'protein sequence'])\n", "('method used for task', 'here , XXXXX , dipeptide composition , correlation features , composition , transition and distribution and pseudo XXXXX are used to represent the XXXXX', ['amino acid composition', 'protein sequence'])\n", "('NONE', 'f1p and f6p can bind to the same XXXXX with different XXXXX', ['active site', 'binding mode'])\n", "('method used for task', 'two XXXXX were used to construct the optimized XXXXX', ['feature selection methods', 'feature subset'])\n", "('NONE', 'our method received a XXXXX using 10-fold cross-validation on the XXXXX', ['high performance', 'training dataset'])\n", "('method used for task', 'XXXXX and XXXXX conformational preferences', ['amino acid', 'secondary structure'])\n", "('method used for task', 'genes connected to the cd4+ t cell subtype XXXXX are more likely to be XXXXX', ['transcription factors', 'master genes'])\n", "('NONE', 'the XXXXX of increased succinate production after gnd gene knockout was consistent with XXXXX', ['experimental data', 'model prediction'])\n", "('method used for task', 'XXXXX for XXXXX can be done in a computationally efficient manner', ['parameter estimation', 'stochastic models'])\n", "('method used for task', 'the strategy combines XXXXX and XXXXX', ['computational modeling', 'experimental analysis'])\n", "('method used for task', 'XXXXX adjustment method ( psam ) is proposed as a tool for XXXXX to improve the power for single locus studies through an estimated XXXXX ( ps )', ['dimension reduction', 'propensity score'])\n", "('NONE', 'molecular interaction and XXXXX of hits were studied by XXXXX', ['binding mode', 'molecular docking'])\n", "('NONE', 'the XXXXX problem can be regarded as a special XXXXX problem in the sense that truly present proteins are those proteins with non-zero abundances', ['protein inference', 'protein quantification'])\n", "('method used for task', 'we test three very simple XXXXX methods to solve the XXXXX problem effectively', ['protein quantification', 'protein inference'])\n", "('method used for task', 'we used the computational strategy which includes XXXXX , secondary structure prediction , comparative modeling and XXXXX ( ppi ) analysis', ['sequence analysis', 'protein–protein interactions'])\n", "('NONE', 'XXXXX reach a stable XXXXX between tsp-1 domain and lskl peptide', ['md simulations', 'binding mode'])\n", "('NONE', 'a constrained XXXXX estimates probabilities in XXXXX', ['hidden markov model', 'linear time'])\n", "('NONE', 'we propose a novel algorithm for learning XXXXX with complete and XXXXX', ['factor analysis', 'incomplete data'])\n", "('method used for task', 'the XXXXX is the varying coefficient fractionally XXXXX', ['regression model', 'exponential model'])\n", "('method used for task', 'estimation of XXXXX for XXXXX is challenging', ['statistical models', 'social networks'])\n", "('method used for task', 'we use XXXXX to impute XXXXX enhancing survival tree analysis', ['bayesian networks', 'missing data'])\n", "('method used for task', 'the XXXXX is learned from XXXXX and used for the imputation', ['bayesian network', 'incomplete data'])\n", "('method used for task', 'a variational XXXXX is developed for XXXXX', ['em algorithm', 'parameter estimation'])\n", "('NONE', 'a new hbic criterion is proposed for XXXXX in XXXXX', ['model selection', 'mixture models'])\n", "('method used for task', 'we introduce apfa as XXXXX for discrete XXXXX', ['graphical models', 'longitudinal data'])\n", "('NONE', 'the lower and upper XXXXX are estimated by XXXXX', ['maximum likelihood', 'tail dependence parameters'])\n", "('method used for task', 'XXXXX and XXXXX show that the new approximations are very accurate for all practical purposes', ['theoretical analysis', 'empirical evaluation'])\n", "('method used for task', 'we derive value at risk and XXXXX for gtl-distributed XXXXX', ['expected shortfall', 'random variables'])\n", "('NONE', 'XXXXX show high XXXXX , low false positive rate', ['simulation studies', 'detection rate'])\n", "('NONE', 'XXXXX was demonstrated in simulations and real XXXXX', ['high performance', 'data analysis'])\n", "('NONE', 'under the pot framework , we estimate tail XXXXX . extensive XXXXX show the new method works well', ['risk measures', 'simulation studies'])\n", "('method used for task', 'XXXXX is constructed based on XXXXX to improve retrieval efficiency', ['case base', 'domain ontology'])\n", "('method used for task', 'the XXXXX is defined based on the XXXXX', ['similarity function', 'cost function'])\n", "('method used for task', 'automatic collection of constraint data from XXXXX for XXXXX', ['cad models', 'assembly planning'])\n", "('method used for task', 'generated XXXXX are validated via interference and XXXXX', ['stability analysis', 'assembly sequences'])\n", "('NONE', 'a schema is defined for XXXXX that represent XXXXX as networks', ['spatial relations', 'space layouts'])\n", "('method used for task', 'XXXXX for spatial XXXXX extend the schema', ['spatial constraints', 'consistency checking'])\n", "('NONE', 'XXXXX of XXXXX and simulation for milling in the field of dental technology', ['integrated approach', 'tool path generation'])\n", "('method used for task', 'a kind of XXXXX is adopted to describe the XXXXX', ['region codes', 'surface regions'])\n", "('NONE', 'we propose a comprehensive XXXXX geometrical XXXXX', ['human body', 'shape representation'])\n", "('NONE', 'XXXXX of recommended XXXXX types is implemented', ['automatic generation', 'assembly tolerance'])\n", "('NONE', 'perfect integration of XXXXX in a XXXXX', ['subdivision surfaces', 'cad system'])\n", "('NONE', 'a XXXXX classifies various designs as three-level XXXXX', ['representation model', 'design elements'])\n", "('method used for task', 'it combines csg modeling and XXXXX based shape and XXXXX', ['level set', 'topology optimization'])\n", "('method used for task', 'feedrate planning under confined XXXXX is reduced to XXXXX', ['tracking error', 'convex optimization'])\n", "('method used for task', 'XXXXX taubin integral for smooth or piecewise XXXXX', ['closed form', 'smooth surfaces'])\n", "('method used for task', 'lspia is so flexible that it allows the adjustment of the number of XXXXX , and a XXXXX in the iterations', ['control points', 'knot vector'])\n", "('NONE', 'numerical behavior of XXXXX through XXXXX', ['singular value decomposition', 'matrix representations'])\n", "('NONE', 'XXXXX of XXXXX and advection–diffusion demonstrated', ['isogeometric analysis', 'linear elasticity'])\n", "('NONE', 'algorithm for computing voxelized XXXXX of two XXXXX', ['minkowski sum', 'input polyhedra'])\n", "('method used for task', 'a new XXXXX for the XXXXX via geometric computations', ['analysis method', 'tolerance estimation'])\n", "('NONE', 'employing two pre-built XXXXX , a hierarchical gauss map and a coons bounding volume hierarchy , we develop an efficient culling technique that can eliminate the majority of redundant XXXXX', ['data structures', 'surface patches'])\n", "('NONE', 'demonstrated by a XXXXX of multiobjective XXXXX of a paper mill', ['case study', 'design optimization'])\n", "('method used for task', 'we model an XXXXX based method for mechanical XXXXX', ['ant colony optimization', 'assembly planning'])\n", "('method used for task', 'we propose a new approach to process eeg and XXXXX for XXXXX', ['hrv signals', 'design activities'])\n", "('NONE', 'family mould XXXXX using XXXXX and mould XXXXX grammars is proposed', ['genetic algorithms', 'layout design'])\n", "('NONE', 'group-oriented mould XXXXX grammar-based XXXXX and chromosome are proposed', ['layout design', 'genetic operators'])\n", "('NONE', 'novel use of the existing XXXXX that scale linearly with the XXXXX', ['data structures', 'problem size'])\n", "('NONE', 'optimal XXXXX of XXXXX with various geometric constraints is presented', ['degree reduction', 'bézier curves'])\n", "('NONE', 'optimal XXXXX of bézier curves with various XXXXX is presented', ['degree reduction', 'geometric constraints'])\n", "('NONE', 'optimal degree reduction of XXXXX with various XXXXX is presented', ['bézier curves', 'geometric constraints'])\n", "('NONE', 'XXXXX shapes are XXXXX representatives from a simulation perspective', ['skin model', 'skin shapes'])\n", "('NONE', 'XXXXX shapes are XXXXX representatives from a simulation perspective', ['skin shapes', 'skin model'])\n", "('method used for task', 'an efficient XXXXX suitable for XXXXX', ['interactive applications', 'parallel solver'])\n", "('method used for task', 'we present a novel bci based XXXXX for conceptual XXXXX', ['user interface', '3d modeling'])\n", "('NONE', 'a XXXXX study to assess intuitiveness of bci in XXXXX is presented', ['human factors', '3d modeling'])\n", "('method used for task', 'we present a method for automatic tool-adaptive XXXXX for XXXXX', ['path planning', 'freeform surfaces'])\n", "('method used for task', 'derivation of XXXXX for loaded XXXXX with a varying isotropic stress', ['shell structures', 'equilibrium equations'])\n", "('method used for task', 'the application of the XXXXX for the XXXXX', ['finite element method', 'numerical implementation'])\n", "('method used for task', 'the introduction of a new XXXXX for the XXXXX', ['particle method', 'numerical implementation'])\n", "('NONE', 'a multi-dimensional XXXXX based XXXXX fitting scheme to a freeform rational surface', ['dynamic programming', 'ruled surface'])\n", "('method used for task', 'highly XXXXX running on gpus are employed to evaluate the multi-dimensional XXXXX', ['parallel algorithms', 'dynamic programming'])\n", "('method used for task', 'unified XXXXX for both static–dynamic load analysis , XXXXX and form finding', ['structural optimization', 'physics engine'])\n", "('NONE', 'an overview is given that combine existing casting techniques with XXXXX for the fabrication of complex XXXXX', ['digital fabrication', 'concrete structures'])\n", "('NONE', 'present a perspective overview of the XXXXX about 3d XXXXX', ['future research', 'tolerance analysis'])\n", "('NONE', 'XXXXX were extracted based on a local score of XXXXX', ['boundary points', 'data points'])\n", "('NONE', 'we investigate the use of XXXXX for the exchange of “intelligent” XXXXX', ['semantic web technologies', 'cad models'])\n", "('method used for task', 'we define axioms and XXXXX to achieve XXXXX between cad ontologies', ['semantic integration', 'mapping rules'])\n", "('NONE', 'we extend XXXXX with a XXXXX to detect similar design features', ['semantic integration', 'similarity measurement'])\n", "('method used for task', 'we extend XXXXX with a similarity measurement to detect similar XXXXX', ['semantic integration', 'design features'])\n", "('method used for task', 'we extend semantic integration with a XXXXX to detect similar XXXXX', ['similarity measurement', 'design features'])\n", "('method used for task', 'the XXXXX is used to create a dual-level XXXXX . numerical instabilities can be improved', ['density approximation', 'shepard function'])\n", "('method used for task', 'a new spiral XXXXX based on space archimedean spiral is proposed to machining XXXXX with big slope', ['tool path generation', 'freeform surfaces'])\n", "('method used for task', 'we present shared XXXXX ( sst ) to rasterize XXXXX on gpus', ['implicit surfaces', 'subdivision trees'])\n", "('method used for task', 'XXXXX is adopted to train the XXXXX accumulatively', ['online learning', 'segmentation model'])\n", "('method used for task', 'we present a XXXXX to compute and update the mesh structure efficiently during XXXXX', ['parallel algorithm', 'image registration'])\n", "('method used for task', 'double tangential contact between the tool and the XXXXX is employed to connect feasible hyper-osculating XXXXX', ['tool paths', 'target surface'])\n", "('NONE', 'we introduce a XXXXX that maximizes the geometric matching between the tool and the XXXXX', ['global optimization algorithm', 'target surface'])\n", "('NONE', 'the algorithm analyzes high-level XXXXX existent in a XXXXX', ['semantic information', '3d model'])\n", "('NONE', 'we reduce the XXXXX dramatically by solving XXXXX', ['computation time', 'convex optimization problem'])\n", "('NONE', 'we propose two intrinsic methods for computing XXXXX ( cvt ) on XXXXX', ['centroidal voronoi tessellation', 'triangle meshes'])\n", "('method used for task', 'models are derived using XXXXX , design geometry and XXXXX', ['simulation intent', 'analysis attributes'])\n", "('NONE', 'by specifying a different XXXXX , different XXXXX are obtained', ['simulation intent', 'analysis models'])\n", "('NONE', 'XXXXX of surface meshes are introduced to XXXXX', ['geometric constraints', 'global optimization'])\n", "('NONE', 'XXXXX of XXXXX are introduced to global optimization', ['geometric constraints', 'surface meshes'])\n", "('method used for task', 'geometric constraints of XXXXX are introduced to XXXXX', ['global optimization', 'surface meshes'])\n", "('method used for task', 'a XXXXX is proposed to describe the local shape of the XXXXX', ['local descriptor', 'interest points'])\n", "('NONE', 'we prove that for a given XXXXX , the XXXXX exists only in very special cases ( for a special form of an angle function )', ['b-spline curve', 'exact solution'])\n", "('NONE', 'we propose an algorithm for finding an XXXXX , derive a bound on its XXXXX and study the approximation order of the proposed algorithm', ['approximate solution', 'approximation error'])\n", "('method used for task', 'XXXXX represents a crucial factor for good XXXXX', ['image quality', 'lens system design'])\n", "('NONE', 'XXXXX is the main part of the XXXXX methodology', ['optimization procedure', 'lens system design'])\n", "('NONE', 'XXXXX is modeled as a network of XXXXX agents', ['design space', 'design parameter'])\n", "('NONE', 'XXXXX agents help to overcome the cognitive limitation of XXXXX', ['design parameter', 'role agents'])\n", "('method used for task', 'a XXXXX is obtained using two different XXXXX', ['confidence interval', 'linearization strategies'])\n", "('NONE', 'an integrated XXXXX for identifying contextual impacts of XXXXX', ['product model', 'design changes'])\n", "('NONE', 'we propose a new approach to the problem of XXXXX of XXXXX', ['degree reduction', 'bézier curves'])\n", "('NONE', 'we devise a cooperative awareness model to describe cooperative XXXXX in XXXXX', ['product design', 'awareness information'])\n", "('method used for task', 'a mechanism which responds to changes of lean cooperative XXXXX is proposed to plan and execute task in XXXXX', ['product design', 'awareness information'])\n", "('method used for task', 'XXXXX has high geometric fidelity and low XXXXX', ['approximation error', 'steiner format'])\n", "('method used for task', 'an XXXXX to generate smooth spiral curves on XXXXX', ['efficient algorithm', 'free-form surfaces'])\n", "('method used for task', 'an XXXXX is given to improve the conformality of XXXXX', ['optimization method', 'nurbs surfaces'])\n", "('method used for task', 'we develop an analytical representation of XXXXX for XXXXX', ['conformal mapping', 'implicit surfaces'])\n", "('NONE', 'the XXXXX of the XXXXX is the description length', ['quality measure', 'statistical shape model'])\n", "('NONE', 'the XXXXX of the statistical shape model is the XXXXX', ['quality measure', 'description length'])\n", "('method used for task', 'the quality measure of the XXXXX is the XXXXX', ['statistical shape model', 'description length'])\n", "('method used for task', 'the approaches ground on algorithms from XXXXX and XXXXX', ['computational geometry', 'computer graphics'])\n", "('method used for task', 'we develop a symbolic XXXXX to construct trivariate solids and carry out XXXXX', ['modeling framework', 'isogeometric analysis'])\n", "('method used for task', 'a XXXXX has been developed for XXXXX of pmi in step files', ['software tool', 'conformance checking'])\n", "('NONE', 'a XXXXX has been developed for conformance checking of pmi in XXXXX', ['software tool', 'step files'])\n", "('NONE', 'a software tool has been developed for XXXXX of pmi in XXXXX', ['conformance checking', 'step files'])\n", "('NONE', 'a novel teeth XXXXX by combining optical scan data and dental XXXXX is introduced', ['modeling framework', 'ct images'])\n", "('method used for task', 'an XXXXX is developed to fill a XXXXX with an all-hex mesh', ['iterative algorithm', 'triangular mesh'])\n", "('NONE', 'examples of application of the strategy for XXXXX in XXXXX and cad', ['adaptive refinement', 'isogeometric analysis'])\n", "('method used for task', 'we propose a grid-free XXXXX for analytic XXXXX', ['geometric modeling', 'discretization scheme'])\n", "('NONE', 'we propose an efficient method for extracting XXXXX from the XXXXX', ['feature lines', 'shape collection'])\n", "('NONE', 'we solve the initial XXXXX on XXXXX by solving a first-order ode', ['triangle meshes', 'value problem'])\n", "('method used for task', 'we propose a method for fitting a g2 quadratic XXXXX to planar styling XXXXX', ['b-spline curve', 'design data'])\n", "('NONE', 'XXXXX is attained by using a non-uniform knot vector of the XXXXX', ['g2 continuity', 'b-spline curve'])\n", "('method used for task', 'XXXXX is attained by using a non-uniform XXXXX of the b-spline curve', ['g2 continuity', 'knot vector'])\n", "('NONE', 'g2 continuity is attained by using a non-uniform XXXXX of the XXXXX', ['b-spline curve', 'knot vector'])\n", "('method used for task', 'adjoint-based XXXXX for XXXXX', ['error estimates', 'euler equations'])\n", "('method used for task', 'an XXXXX for smoothing the XXXXX is proposed', ['iterative procedure', 'frame field'])\n", "('NONE', 'XXXXX using only XXXXX as inputs was developed', ['user interface', 'cad models'])\n", "('NONE', 'b-age derives a XXXXX or XXXXX only from salient 3d points', ['b-spline curve', 'surface estimation'])\n", "('method used for task', 'average strip width XXXXX and sample points selection method for 3+2-axis XXXXX', ['sculptured surface machining', 'estimation method'])\n", "('NONE', 'novel link between XXXXX ( discrete XXXXX ) and cagd ( smooth patchworks from bezier surfaces of degree ( 1,1 ) )', ['discrete differential geometry', 'affine minimal surfaces'])\n", "('method used for task', 'a XXXXX to discrete XXXXX , based on smooth patchworks', ['geometric approach', 'affine minimal surfaces'])\n", "('method used for task', 'a XXXXX to compute a μ-basis for complex XXXXX', ['fast algorithm', 'rational curves'])\n", "('method used for task', 'main properties and several applications from dynamics , XXXXX and geometry , XXXXX on polyhedra , motion of rigid body and positive definite matrices', ['medical imaging', 'curve design'])\n", "('NONE', 'main properties and several applications from dynamics , XXXXX and geometry , curve design on polyhedra , motion of XXXXX and positive definite matrices', ['medical imaging', 'rigid body'])\n", "('NONE', 'main properties and several applications from dynamics , medical imaging and geometry , XXXXX on polyhedra , motion of XXXXX and positive definite matrices', ['curve design', 'rigid body'])\n", "('method used for task', 'a novel XXXXX method is proposed to compute the intersection between a ray and a XXXXX using the XXXXX approximation', ['parametric surface', 'second order'])\n", "('method used for task', 'it is a geometric XXXXX which is less sensitive to XXXXX than newton–raphson and halley methods', ['initial conditions', 'iteration scheme'])\n", "('method used for task', 'modified t-splines are favorable both in adaptive XXXXX and XXXXX', ['geometric modeling', 'isogeometric analysis'])\n", "('method used for task', 'it is proven that linear XXXXX lead to reparameterized cubic XXXXX', ['quaternion polynomials', 'ph curves'])\n", "('NONE', 'spatial rational XXXXX of a class m=3,4,5,6 are derived in a XXXXX having 2m+4 degrees of freedom', ['closed form', 'ph curves'])\n", "('method used for task', 'our XXXXX are both monotonicity preserving and XXXXX', ['convexity preserving', 'limit functions'])\n", "('NONE', 'an XXXXX for bound of the XXXXX is designed', ['hausdorff distance', 'estimation method'])\n", "('NONE', 'efficient evaluation through a XXXXX of the XXXXX', ['linear transformation', 'control polygon'])\n", "('NONE', 'the scheme produces XXXXX spatial XXXXX in 3d', ['convexity preserving', 'limit curves'])\n", "('method used for task', 'the XXXXX are proved to be g1 continuous , while XXXXX show that they are also g2 smooth and fair', ['numerical examples', 'limit curves'])\n", "('NONE', 'several examples , involving the XXXXX parametric speed , XXXXX , tangent , normal , curvature , and offsets are presented', ['arc length', 'sweep curve'])\n", "('method used for task', 'the initial parameters of the nurbs-snake are obtained using a two step procedure in which firstly weights are obtained by solving a XXXXX and then XXXXX are evolved by solving a system of linear equation', ['control points', 'quadratic programming problem'])\n", "('NONE', 'the conditions for XXXXX between toric XXXXX are analyzed', ['g1 continuity', 'surface patches'])\n", "('method used for task', 'some practical XXXXX for XXXXX are developed', ['sufficient conditions', 'g1 continuity'])\n", "('NONE', 'the proposed methods allow existing XXXXX to fully exploit the advantageous properties of XXXXX within the context of prevailing cad geometry representations', ['cad systems', 'ph curves'])\n", "('NONE', 'new XXXXX of XXXXX', ['direct method', 'curve design'])\n", "('method used for task', 'we construct univariate XXXXX for XXXXX', ['subdivision schemes', 'noisy data'])\n", "('method used for task', 'a XXXXX is analyzed and validated by several XXXXX', ['statistical model', 'numerical examples'])\n", "('NONE', 'XXXXX of the algorithm adapts to the geometry of XXXXX', ['step size', 'parametric surface'])\n", "('method used for task', 'we derive a new XXXXX that is equal to the known XXXXX in most cases', ['upper bound', 'lower bound'])\n", "('method used for task', 'we devise a XXXXX for XXXXX based on rdf', ['data-driven approach', 'object modeling'])\n", "('NONE', 'we introduce XXXXX into XXXXX and co-segmentation', ['deep learning', '3d shape segmentation'])\n", "('NONE', 'the proposed scheme does not require any XXXXX of XXXXX', ['prior knowledge', 'network topology'])\n", "('NONE', 'ralp’s main disadvantage is its XXXXX as the message exchange rounds depend on the XXXXX . in this article , we introduce the concept of clusters , where the data transfer can proceed after a certain number of message exchange rounds', ['convergence time', 'neighborhood graph'])\n", "('NONE', 'ralp’s main disadvantage is its XXXXX as the message exchange rounds depend on the neighborhood graph . in this article , we introduce the concept of clusters , where the XXXXX can proceed after a certain number of message exchange rounds', ['convergence time', 'data transfer'])\n", "('NONE', 'ralp’s main disadvantage is its XXXXX as the XXXXX rounds depend on the neighborhood graph . in this article , we introduce the concept of clusters , where the data transfer can proceed after a certain number of XXXXX rounds', ['convergence time', 'message exchange'])\n", "('NONE', 'ralp’s main disadvantage is its XXXXX as the XXXXX rounds depend on the neighborhood graph . in this article , we introduce the concept of clusters , where the data transfer can proceed after a certain number of XXXXX rounds', ['convergence time', 'message exchange'])\n", "('NONE', 'ralp’s main disadvantage is its convergence time as the message exchange rounds depend on the XXXXX . in this article , we introduce the concept of clusters , where the XXXXX can proceed after a certain number of message exchange rounds', ['neighborhood graph', 'data transfer'])\n", "('method used for task', 'ralp’s main disadvantage is its convergence time as the XXXXX rounds depend on the XXXXX . in this article , we introduce the concept of clusters , where the data transfer can proceed after a certain number of XXXXX rounds', ['neighborhood graph', 'message exchange'])\n", "('method used for task', 'ralp’s main disadvantage is its convergence time as the XXXXX rounds depend on the XXXXX . in this article , we introduce the concept of clusters , where the data transfer can proceed after a certain number of XXXXX rounds', ['neighborhood graph', 'message exchange'])\n", "('NONE', 'ralp’s main disadvantage is its convergence time as the XXXXX rounds depend on the neighborhood graph . in this article , we introduce the concept of clusters , where the XXXXX can proceed after a certain number of XXXXX rounds', ['data transfer', 'message exchange'])\n", "('NONE', 'ralp’s main disadvantage is its convergence time as the XXXXX rounds depend on the neighborhood graph . in this article , we introduce the concept of clusters , where the XXXXX can proceed after a certain number of XXXXX rounds', ['data transfer', 'message exchange'])\n", "('NONE', 'proposing a new approach to XXXXX in XXXXX', ['barrier coverage', 'wireless sensor network'])\n", "('NONE', 'cause-effect relationships between service qos metrics and vn XXXXX being modelled as XXXXX', ['bayesian network', 'stress states'])\n", "('method used for task', 'concise and up-to-date review of XXXXX for XXXXX', ['quality assessment', 'video streaming services'])\n", "('method used for task', 'discussion of XXXXX and challenges for qoe in XXXXX', ['future trends', 'video streaming services'])\n", "('NONE', 'we present an integrated platform that considers XXXXX of XXXXX', ['social aspects', 'mobile sensing'])\n", "('method used for task', 'we prove the np-hardness of the problem and develop a XXXXX to minimize the XXXXX', ['heuristic algorithm', 'energy consumption'])\n", "('NONE', 'this finding allowed for positive verification of the social XXXXX in XXXXX', ['online communities', 'exchange theory'])\n", "('method used for task', 'we show how twitter XXXXX relate to XXXXX pricing', ['volume spikes', 'stock options'])\n", "('method used for task', 'we provide a XXXXX to solve policy-aware XXXXX', ['heuristic algorithm', 'peering problem'])\n", "('NONE', 'we examine the XXXXX of peering connections to cover XXXXX in today’s internet', ['lower bound', 'end users'])\n", "('method used for task', 'large-scale computational generation and motif XXXXX for the synthetic XXXXX . we obtain a distinct motif pattern for each such class', ['distribution analysis', 'topology classes'])\n", "('NONE', 'comprehensive XXXXX of XXXXX ( facebook , twitter , google plus ) from which we obtain three quantifiable characteristic motif fingerprints', ['online social networks', 'motif analysis'])\n", "('method used for task', 'we propose robust access control framework for a network which has allowed XXXXX to be connected to the internal network in order to enable seamless XXXXX', ['smart devices', 'data sharing'])\n", "('method used for task', 'smart device’s XXXXX such as location , app usage pattern , unlock failures are being considered for XXXXX and data confidentiality', ['sensor data', 'access control'])\n", "('method used for task', 'smart device’s XXXXX such as location , app usage pattern , unlock failures are being considered for access control and XXXXX', ['sensor data', 'data confidentiality'])\n", "('method used for task', 'smart device’s sensor data such as location , app usage pattern , unlock failures are being considered for XXXXX and XXXXX', ['access control', 'data confidentiality'])\n", "('method used for task', 'the algorithm supports both the XXXXX and XXXXX simultaneously', ['access control', 'data confidentiality'])\n", "('NONE', 'XXXXX are effectiveness , efficiency , XXXXX , and fairness', ['performance metrics', 'power consumption'])\n", "('NONE', 'to cope with large-scale XXXXX , we introduce techniques to make the sfmap framework scalable . we validate the effectiveness of the approach using large-scale XXXXX collected at a gateway point of internet access links', ['measurement data', 'traffic data'])\n", "('method used for task', 'XXXXX are a good choice for their XXXXX and low energy consumption', ['decision trees', 'high accuracy'])\n", "('NONE', 'barometers offer XXXXX , XXXXX and independence from position', ['high accuracy', 'energy efficiency'])\n", "('method used for task', 'a XXXXX based on henkin models is given for a predicative polymorphic calculus of XXXXX', ['denotational semantics', 'access control'])\n", "('NONE', 'several XXXXX included an extension of aspectj that supports the new XXXXX', ['case studies', 'annotation model'])\n", "('method used for task', 'presents two XXXXX for a+i programs and proves that both are fully abstract w.r.t . the XXXXX', ['trace semantics', 'operational semantics'])\n", "('NONE', 'we discuss a XXXXX of the moldable XXXXX', ['prototype implementation', 'debugger model'])\n", "('NONE', 'we establish a XXXXX that captures the XXXXX of language workbenches', ['feature model', 'design space'])\n", "('NONE', 'tao is the first attempt to apply the methodology of XXXXX in test and XXXXX', ['denotational semantics', 'oracle generation'])\n", "('method used for task', 'we analyze the problem and XXXXX for XXXXX in c++', ['design space', 'actor programming'])\n", "('method used for task', 'the XXXXX is an extension to the communicating event-loop XXXXX', ['domain model', 'actor model'])\n", "('NONE', 'the XXXXX retains the safety and liveness properties of the XXXXX', ['domain model', 'actor model'])\n", "('NONE', 'we provide an XXXXX and validation of the XXXXX', ['operational semantics', 'domain model'])\n", "('NONE', 'mcerlang is used for XXXXX of timed XXXXX based on monitors', ['model checking', 'rebeca models'])\n", "('method used for task', 'the existing XXXXX is extended to calculate the XXXXX for simulation results', ['simulation tool', 'confidence interval'])\n", "('method used for task', 'bayesian belief networks ( bbns ) and XXXXX ( fcms ) , as dynamic influence graphs , were applied to handle the task of medical knowledge formalization for XXXXX', ['fuzzy cognitive maps', 'decision support'])\n", "('NONE', 'for reasoning on these knowledge models , a XXXXX reasoning engine , eye , with the necessary plug-ins was developed in the XXXXX', ['general purpose', 'semantic web'])\n", "('method used for task', 'developing a XXXXX for constructing a network from a XXXXX in linear time', ['fast algorithm', 'time series'])\n", "('method used for task', 'developing a XXXXX for constructing a network from a time series in XXXXX', ['fast algorithm', 'linear time'])\n", "('NONE', 'developing a fast algorithm for constructing a network from a XXXXX in XXXXX', ['time series', 'linear time'])\n", "('NONE', 'a simulated XXXXX was used to elicit different levels of XXXXX', ['mental workload', 'process control task'])\n", "('method used for task', 'development of systematic XXXXX and knowledge aggregation methodology based on XXXXX', ['knowledge acquisition', 'staff interviews'])\n", "('method used for task', 'use of rad based XXXXX for representing process knowledge aggregated from XXXXX', ['process models', 'staff interviews'])\n", "('NONE', 'the proposed class of flexible weight functions allows emphasis on different XXXXX when comparing XXXXX between two groups', ['time periods', 'cumulative incidence functions'])\n", "('method used for task', 'a large improvement in power was shown when an early XXXXX was used to compare two XXXXX with an early difference', ['weight function', 'cumulative incidence functions'])\n", "('method used for task', 'the methodology can impose constraints on all different compartments , even if XXXXX measurements are not possible by augmenting the XXXXX with an observer', ['control system', 'drug concentration'])\n", "('NONE', 'it is possible to assess potential XXXXX in healthy subjects using XXXXX', ['cardiovascular risk', 'cluster analysis'])\n", "('method used for task', 'a fully automatic XXXXX is proposed for regional classification of the left ventricular wall in XXXXX of small animals', ['ultrasound images', 'processing pipeline'])\n", "('method used for task', 'the pipeline is implemented using state-of-the-art methods from XXXXX and XXXXX', ['computer vision', 'pattern classification'])\n", "('method used for task', 'good performance of the XXXXX is demonstrated on ultrasound data of living mice before and after artificially induced XXXXX', ['myocardial infarction', 'processing pipeline'])\n", "('method used for task', 'we achieve notable improvement in sperm XXXXX and fewer XXXXX compared to the state-of-the-art method', ['false positives', 'head detection'])\n", "('NONE', 'we design and built a XXXXX for the treatment of XXXXX based on asbru', ['decision support system', 'breast cancer'])\n", "('method used for task', 'the XXXXX is improved using XXXXX of vessel structure', ['computational efficiency', 'prior knowledge'])\n", "('NONE', 'we develop an integrated XXXXX ( called ihautisis ) for improving the work efficiency of XXXXX', ['surveillance system', 'infection control'])\n", "('NONE', 'we developed a method for XXXXX of fovea in XXXXX based on vessel-free zone and adaptive gaussian template', ['automated detection', 'fundus images'])\n", "('NONE', 'a semi-online XXXXX in the XXXXX is studied', ['patient scheduling problem', 'pathology laboratory'])\n", "('NONE', 'the approach reduces XXXXX of patients and improves XXXXX', ['waiting time', 'operations efficiency'])\n", "('method used for task', 'by applying XXXXX and support vector machine learning methods to separate pd from hc , we demonstrated 85 % XXXXX', ['classification accuracy', 'feature selection algorithms'])\n", "('method used for task', 'XXXXX based on handwriting analysis can be complementary method to diagnosis made by clinician or other XXXXX', ['decision support system', 'decision support systems'])\n", "('NONE', 'we used three XXXXX to improve the performance of the XXXXX', ['svm classifier', 'feature selection methods'])\n", "('NONE', '13 signal XXXXX were derived from segments of XXXXX', ['quality metrics', 'ecg waveforms'])\n", "('method used for task', 'an XXXXX for fully automatic XXXXX', ['embedded system', 'fall detection'])\n", "('NONE', 'a freely available database for evaluation of XXXXX , consisting of both accelerometric and XXXXX', ['fall detection', 'depth data'])\n", "('method used for task', 'our scheme is based on extended XXXXX and suitable for XXXXX in tmis', ['chaotic maps', 'data exchange'])\n", "('NONE', 'respiratory-gating of 18f-fdg XXXXX improves the accuracy of XXXXX', ['pet images', 'volume estimates'])\n", "('method used for task', 'reduction in XXXXX index ( bmi ) , body fat percentage , and XXXXX', ['body mass', 'body weight loss'])\n", "('NONE', 'XXXXX characteristics affect XXXXX', ['growth plate', 'soc development'])\n", "('NONE', 'the discriminant performance in this study , evaluated as az ( area under the XXXXX ) , for detecting the existence of endotracheal tubes was 0.943±0.009 , and the detection error of the XXXXX was 1.89±2.01mm', ['roc curve', 'tip location'])\n", "('NONE', 'couples the XXXXX of photon transport with XXXXX', ['monte carlo method', 'heat transfer'])\n", "('NONE', 'the new features can enhance the performance of XXXXX that discriminate between malignant and benign XXXXX', ['classification algorithms', 'skin lesions'])\n", "('NONE', 'validation on XXXXX as well as uva XXXXX', ['clinical data', 'simulation data'])\n", "('NONE', 'we propose valworkbench and describe its internal modules and classes , in order to provide the much needed XXXXX for the development and testing of new validation measures as well as XXXXX', ['software platform', 'clustering algorithms'])\n", "('NONE', 'the proposed model combines the demographic XXXXX with the findings of the initial screening mammogram to elicit the hidden and impeding risk of developing XXXXX', ['risk factors', 'breast cancer'])\n", "('method used for task', 'a novel scheme for extracting XXXXX based on morphological XXXXX ( mca ) algorithm is presented in this paper', ['retinal blood vessels', 'component analysis'])\n", "('method used for task', 'a parallel XXXXX ( pgica ) on gpu was proposed for XXXXX', ['group ica', 'fmri data analysis'])\n", "('method used for task', 'provides comprehensive genome scale analysis for users with XXXXX ( hpc ) and gene-level analysis for memory efficient XXXXX', ['high performance computing', 'personal computers'])\n", "('NONE', 'we proved the XXXXX of the XXXXX proposed', ['high accuracy', 'segmentation method'])\n", "('NONE', 'this paper presents an algorithm for registration of XXXXX using orthogonal XXXXX as features', ['retinal images', 'moment invariants'])\n", "('method used for task', 'the trs in test XXXXX with respect to reference XXXXX is estimated using similarity transformation', ['retinal image', 'test retinal image'])\n", "('method used for task', 'the test XXXXX is aligned with reference XXXXX using the estimated registration parameters', ['retinal image', 'test retinal image'])\n", "('method used for task', 'XXXXX was proposed for accurate XXXXX regardless of any variations in the environment loading', ['motion control', 'compensation method'])\n", "('NONE', 'we predicted the outcome of stroke using knowledge discovery process ( kdp ) methods , XXXXX ( ann ) and XXXXX ( svm ) models', ['artificial neural networks', 'support vector machine'])\n", "('method used for task', 'current XXXXX systems fail to transfer a XXXXX to the next available nurse , if the primary nurse is on the phone with another patient', ['nurse call', 'nurse systems'])\n", "('method used for task', 'current XXXXX systems fail to transfer a XXXXX to the next available nurse , if the primary nurse is on the phone with another patient', ['nurse systems', 'nurse call'])\n", "('method used for task', 'current XXXXX systems fail to transfer a XXXXX to the next available nurse , if the primary nurse is currently on another visit', ['nurse call', 'nurse systems'])\n", "('method used for task', 'current XXXXX systems fail to transfer a XXXXX to the next available nurse , if the primary nurse is currently on another visit', ['nurse systems', 'nurse call'])\n", "('method used for task', 'patient-specific XXXXX is necessary to obtain accurate XXXXX', ['cardiac output', 'wall shear stress'])\n", "('NONE', 'a system that provides a convenient way for physicians to retrieve and compare XXXXX among XXXXX providers about herniorrhaphy', ['clinical pathways', 'health care'])\n", "('NONE', 'a cross-correlation function and an average deviation between the continuous XXXXX and the interpolation of limited XXXXX samples are calculated', ['blood glucose', 'blood glucose samples'])\n", "('NONE', 'XXXXX ( emr ) can support a secure , real-time , point-of-care , patient centric XXXXX for clinical care', ['electronic medical record', 'information resource'])\n", "('NONE', '∼14 % lower fpf than XXXXX , bayesian , knn , XXXXX using the same two features', ['level set', 'svm classification'])\n", "('method used for task', 'wind-driven XXXXX ( wso ) is used for optimizing the XXXXX of a clinical decision support system', ['swarm optimization', 'rule base'])\n", "('method used for task', 'wind-driven XXXXX ( wso ) is used for optimizing the rule base of a XXXXX', ['swarm optimization', 'clinical decision support system'])\n", "('NONE', 'wind-driven swarm optimization ( wso ) is used for optimizing the XXXXX of a XXXXX', ['rule base', 'clinical decision support system'])\n", "('NONE', 'XXXXX are utilized as XXXXX', ['shape features', 'adjacency statistics'])\n", "('method used for task', 'we propose 3d XXXXX and tracking algorithm based on an XXXXX and a kalman filter', ['vessel segmentation', 'active contour model'])\n", "('method used for task', 'we propose 3d XXXXX and tracking algorithm based on an active contour model and a XXXXX', ['vessel segmentation', 'kalman filter'])\n", "('method used for task', 'we propose 3d vessel segmentation and tracking algorithm based on an XXXXX and a XXXXX', ['active contour model', 'kalman filter'])\n", "('method used for task', 'the proposed methodology concentrates its efforts on personalized/individual XXXXX including the family history and the demographic XXXXX to elicit the hidden risk of developing breast cancer', ['decision making', 'risk factors'])\n", "('method used for task', 'the proposed methodology concentrates its efforts on personalized/individual XXXXX including the family history and the demographic risk factors to elicit the hidden risk of developing XXXXX', ['decision making', 'breast cancer'])\n", "('NONE', 'the proposed methodology concentrates its efforts on personalized/individual XXXXX including the XXXXX and the demographic risk factors to elicit the hidden risk of developing breast cancer', ['decision making', 'family history'])\n", "('NONE', 'the proposed methodology concentrates its efforts on personalized/individual decision making including the family history and the demographic XXXXX to elicit the hidden risk of developing XXXXX', ['risk factors', 'breast cancer'])\n", "('method used for task', 'the proposed methodology concentrates its efforts on personalized/individual decision making including the XXXXX and the demographic XXXXX to elicit the hidden risk of developing breast cancer', ['risk factors', 'family history'])\n", "('method used for task', 'the proposed methodology concentrates its efforts on personalized/individual decision making including the XXXXX and the demographic risk factors to elicit the hidden risk of developing XXXXX', ['breast cancer', 'family history'])\n", "('method used for task', 'XXXXX on the simulated data to construct a XXXXX', ['predictive model', 'machine learning methods'])\n", "('method used for task', 'new XXXXX based on pso and outlier rejection with XXXXX', ['level set', 'image segmentation method'])\n", "('NONE', 'the use of XXXXX allows an optimal choice of XXXXX', ['pso algorithm', 'cluster centers'])\n", "('NONE', 'XXXXX based segmentation of XXXXX and cup is implemented', ['adaptive threshold', 'optic disc'])\n", "('NONE', 'investigating the XXXXX of long-term hrv for XXXXX in chf patients', ['discrimination power', 'risk assessment'])\n", "('method used for task', 'ecg , XXXXX , respiration , strides and gait XXXXX were measured', ['skin conductance', 'acceleration signals'])\n", "('NONE', 'XXXXX was combined with the kpa-weighted XXXXX as a feature to reduce the influence of noises far from tumor', ['spatial correlation', 'color information'])\n", "('method used for task', 'the proposed approach combined the neutrosophic XXXXX with level set algorithm for XXXXX', ['image segmentation', 'similarity score'])\n", "('NONE', 'neutrosophic similarity score is employed to deal with XXXXX on XXXXX', ['uncertain information', 'image segmentation'])\n", "('method used for task', 'neutrosophic XXXXX is employed to deal with XXXXX on image segmentation', ['uncertain information', 'similarity score'])\n", "('method used for task', 'neutrosophic XXXXX is employed to deal with uncertain information on XXXXX', ['image segmentation', 'similarity score'])\n", "('NONE', 'the results showed that we obtain nature enhanced XXXXX using our method and the XXXXX is very small', ['x-ray images', 'processing time'])\n", "('method used for task', 'XXXXX thermal contrast magnitude depends on the XXXXX and depth as well as on the breast density', ['steady state', 'tumour diameter'])\n", "('method used for task', 'transient thermal contrast trends present three important characteristics including the XXXXX , the transient peak and its corresponding XXXXX', ['response time', 'observation time'])\n", "('NONE', 'a stable XXXXX with suitable force and XXXXX is achieved', ['human–robot interaction', 'path tracking'])\n", "('method used for task', 'corroborating on the appliance of XXXXX for XXXXX', ['classification algorithms', 'image segmentation'])\n", "('method used for task', 'proposing a unified and fully XXXXX for both epicardial and mediastinal fats on cardiac XXXXX', ['ct images', 'automatic segmentation method'])\n", "('method used for task', 'a parameter-dependent analysis of XXXXX is developed through a specific XXXXX', ['articular cartilage', 'computational model'])\n", "('method used for task', 'this XXXXX establishes a basic guide the essential properties to be included into XXXXX of articular cartilage', ['parametric study', 'computational modeling'])\n", "('NONE', 'this XXXXX establishes a basic guide the essential properties to be included into computational modeling of XXXXX', ['parametric study', 'articular cartilage'])\n", "('NONE', 'this parametric study establishes a basic guide the essential properties to be included into XXXXX of XXXXX', ['computational modeling', 'articular cartilage'])\n", "('method used for task', 'this paper presents an XXXXX for the XXXXX of mammograms to assist radiologists in confirming their diagnosis in mammography screening', ['integrated system', 'automatic analysis'])\n", "('method used for task', 'we design a semantic cdss to enable data interoperability and XXXXX for patient-specific XXXXX', ['knowledge sharing', 'clinical decision support'])\n", "('method used for task', 'a constructive XXXXX ( cgp ) is proposed for XXXXX', ['genetic programming', 'epileptic seizure detection'])\n", "('NONE', 'we developed a multi-period XXXXX using the XXXXX', ['similarity measure', 'medical diagnosis method'])\n", "('NONE', 'robust performance without collecting XXXXX or retuning the XXXXX', ['experimental data', 'control parameters'])\n", "('NONE', 'optimize the XXXXX by a modified alternative XXXXX', ['objective function', 'iterative algorithm'])\n", "('method used for task', 'XXXXX is exploited for combining XXXXX to improve the performance', ['ensemble learning', 'multiple models'])\n", "('NONE', 'in this work glaucoma identification is done using XXXXX of XXXXX', ['optic disk', 'wavelet features'])\n", "('NONE', 'wavelet features are extracted from segmented and XXXXX removed XXXXX', ['blood vessels', 'optic disk'])\n", "('method used for task', 'XXXXX are extracted from segmented and XXXXX removed optic disk', ['blood vessels', 'wavelet features'])\n", "('method used for task', 'XXXXX are extracted from segmented and blood vessels removed XXXXX', ['optic disk', 'wavelet features'])\n", "('method used for task', 'several XXXXX are used for prominent XXXXX', ['machine learning algorithms', 'feature selection'])\n", "('method used for task', 'XXXXX is used to reduce the dimensionality of XXXXX', ['genetic algorithm', 'feature vector'])\n", "('method used for task', 'automated segmentation of XXXXX for skeletal XXXXX assessment', ['x-ray images', 'bone age'])\n", "('method used for task', 'a XXXXX was applied to track the endocardial XXXXX', ['non-rigid registration', 'boundary points'])\n", "('NONE', 'implement a single prospective XXXXX or XXXXX of phase i combination trials in oncology', ['clinical trial', 'simulation studies'])\n", "('NONE', 'a wireless XXXXX measuring device used together with a smart XXXXX was developed', ['blood pressure', 'mobile device'])\n", "('NONE', 'found the XXXXX of XXXXX for type ii diabetic patients', ['risk factors', 'liver cancer'])\n", "('method used for task', 'built a XXXXX for XXXXX prediction for diabetic patient', ['web-based application', 'liver cancer'])\n", "('method used for task', 'bp XXXXX ( bpann ) is improved by XXXXX ( ga ) and levenberg–marquardt ( lm ) algorithm to model the relation among principal components ( pcs ) and brain age', ['artificial neural network', 'hybrid genetic algorithm'])\n", "('method used for task', 'bp XXXXX ( bpann ) is improved by hybrid genetic algorithm ( ga ) and levenberg–marquardt ( lm ) algorithm to model the relation among XXXXX ( pcs ) and brain age', ['artificial neural network', 'principal components'])\n", "('method used for task', 'bp XXXXX ( bpann ) is improved by hybrid genetic algorithm ( ga ) and levenberg–marquardt ( lm ) algorithm to model the relation among principal components ( pcs ) and XXXXX', ['artificial neural network', 'brain age'])\n", "('method used for task', 'bp artificial neural network ( bpann ) is improved by XXXXX ( ga ) and levenberg–marquardt ( lm ) algorithm to model the relation among XXXXX ( pcs ) and brain age', ['hybrid genetic algorithm', 'principal components'])\n", "('method used for task', 'bp artificial neural network ( bpann ) is improved by XXXXX ( ga ) and levenberg–marquardt ( lm ) algorithm to model the relation among principal components ( pcs ) and XXXXX', ['hybrid genetic algorithm', 'brain age'])\n", "('method used for task', 'bp artificial neural network ( bpann ) is improved by hybrid genetic algorithm ( ga ) and levenberg–marquardt ( lm ) algorithm to model the relation among XXXXX ( pcs ) and XXXXX', ['principal components', 'brain age'])\n", "('NONE', 'analytical formulas for XXXXX , XXXXX , and flow rate are given', ['velocity profile', 'wall shear stress'])\n", "('method used for task', 'analytical formulas for XXXXX , wall shear stress , and XXXXX are given', ['velocity profile', 'flow rate'])\n", "('method used for task', 'analytical formulas for velocity profile , XXXXX , and XXXXX are given', ['wall shear stress', 'flow rate'])\n", "('NONE', 'a review about 3d XXXXX of pulmonary nodules in XXXXX is presented', ['automatic detection', 'ct images'])\n", "('NONE', 'in emergency and crisis situations , many XXXXX can be unserviceable because of damage to equipment or loss of power . thus , XXXXX over wireless communication to achieve uninterrupted network services is a major obstacle', ['data transmission', 'communication channels'])\n", "('method used for task', 'the proposed middleware was ported into an XXXXX , which is compatible with the actual network environment without the need for changing the original XXXXX', ['embedded system', 'system architecture'])\n", "('method used for task', 'the proposed middleware was ported into an XXXXX , which is compatible with the actual XXXXX without the need for changing the original system architecture', ['embedded system', 'network environment'])\n", "('method used for task', 'the proposed middleware was ported into an embedded system , which is compatible with the actual XXXXX without the need for changing the original XXXXX', ['system architecture', 'network environment'])\n", "('NONE', 'method focused on mass-preserving registration of XXXXX with XXXXX', ['large deformation', 'lung images'])\n", "('NONE', 'the work proposes a novel XXXXX that provides a set of methods for metabolomics and spectral XXXXX', ['r package', 'data analysis'])\n", "('NONE', 'the provided functions include data loading , pre-processing , metabolite identification , univariate/multivariate XXXXX , XXXXX and feature selection', ['data analysis', 'machine learning'])\n", "('method used for task', 'the provided functions include data loading , pre-processing , metabolite identification , univariate/multivariate XXXXX , machine learning and XXXXX', ['data analysis', 'feature selection'])\n", "('method used for task', 'the provided functions include data loading , pre-processing , metabolite identification , univariate/multivariate data analysis , XXXXX and XXXXX', ['machine learning', 'feature selection'])\n", "('NONE', 'the biomedical XXXXX aims to speed up the construction of biomedical domain-specific XXXXX', ['search engines', 'search engine framework'])\n", "('NONE', 'we made the implementation of hubness-aware XXXXX available in the pyhubs XXXXX', ['machine learning techniques', 'software package'])\n", "('NONE', 'XXXXX of systems with different XXXXX', ['comparative analysis', 'feature sets'])\n", "('NONE', 'understanding the XXXXX of the XXXXX', ['reliability analysis', 'cadx system'])\n", "('method used for task', 'four methods are tested : a perceptron multilayer , self-organising maps , a XXXXX and XXXXX', ['radial basis function neural network', 'decision trees'])\n", "('NONE', \"glomerulus diameter and bowman 's XXXXX in renal XXXXX indicate various diseases\", ['capsule thickness', 'microscopic images'])\n", "('method used for task', \"this work proposed the analysis particles algorithm based on median filter for morphological image processing to detect the renal corpuscle objects . afterwards , the XXXXX and bowman 's XXXXX are measured\", ['glomerulus diameter', 'capsule thickness'])\n", "('method used for task', 'we used the XXXXX model method on a XXXXX of the wrist joint', ['3d model', 'rigid body spring'])\n", "('NONE', 'we used the rigid body spring model method on a XXXXX of the XXXXX', ['3d model', 'wrist joint'])\n", "('NONE', 'we used the XXXXX model method on a 3d model of the XXXXX', ['rigid body spring', 'wrist joint'])\n", "('NONE', 'data obtained from earlier scans are used to construct a XXXXX with laplacian XXXXX', ['probabilistic atlas', 'mixture model'])\n", "('NONE', 'XXXXX obtained from a XXXXX is modeled for the ct image reconstruction', ['prior information', 'probabilistic atlas'])\n", "('NONE', 'mean , XXXXX , and root-mean-squared value parameters of the signal XXXXX are computed to characterize the signal fluctuations', ['standard deviation', 'envelope amplitude'])\n", "('method used for task', 'an epidemiologic system analysis of XXXXX is presented through a XXXXX', ['cardiovascular risk', 'bayesian network model'])\n", "('method used for task', 'the induced XXXXX was used to make inferences taking into account three reasoning patterns : XXXXX , evidential reasoning , and intercausal reasoning', ['bayesian network', 'causal reasoning'])\n", "('method used for task', 'XXXXX ( pe ) is a XXXXX to evaluate the irregularity of signals', ['permutation entropy', 'fast method'])\n", "('method used for task', 'a hybrid XXXXX proposed with svd and ezw for XXXXX', ['ecg signals', 'compression method'])\n", "('NONE', 'a framework including XXXXX , XXXXX and annotation combination stages for a confidence based benchmark dataset for retinal image processing is proposed', ['data selection', 'task assignment'])\n", "('method used for task', 'a framework including XXXXX , task assignment and annotation combination stages for a confidence based XXXXX for retinal image processing is proposed', ['data selection', 'benchmark dataset'])\n", "('method used for task', 'a framework including data selection , XXXXX and annotation combination stages for a confidence based XXXXX for retinal image processing is proposed', ['task assignment', 'benchmark dataset'])\n", "('method used for task', 'the framework is used to build a confidence based XXXXX for XXXXX', ['benchmark dataset', 'cyst segmentation'])\n", "('NONE', 'the main objective of this study was to investigate the impact of implementing a computer-based order entry system without XXXXX on the number of radiographs ordered for patients seen in the XXXXX', ['clinical decision support', 'emergency department'])\n", "('NONE', 'the main objective of this study was to investigate the impact of implementing a computer-based XXXXX system without XXXXX on the number of radiographs ordered for patients seen in the emergency department', ['clinical decision support', 'order entry'])\n", "('NONE', 'the main objective of this study was to investigate the impact of implementing a computer-based XXXXX system without clinical decision support on the number of radiographs ordered for patients seen in the XXXXX', ['emergency department', 'order entry'])\n", "('NONE', 'our results show a decrease in the number of radiographs ordered after computer-based XXXXX system implementation , despite an increase in the number of XXXXX admissions', ['emergency department', 'order entry'])\n", "('NONE', 'our study also shows that the XXXXX between XXXXX admission and medical imaging was not affected by this new workflow', ['time interval', 'emergency department'])\n", "('method used for task', 'our study also shows that the XXXXX between emergency department admission and XXXXX was not affected by this new workflow', ['time interval', 'medical imaging'])\n", "('method used for task', 'our study also shows that the time interval between XXXXX admission and XXXXX was not affected by this new workflow', ['emergency department', 'medical imaging'])\n", "('method used for task', 'our proposed XXXXX score is similar to the XXXXX of glaucoma experts', ['quality assessment', 'quality index'])\n", "('NONE', 'a XXXXX of XXXXX ( tf ) was proposed by a tlm model of human artery tree', ['transfer function', 'calculation method'])\n", "('method used for task', 'this study proposed and examined a new approach employing the light sharing pet XXXXX with thick light guide and gapd array having large-area microcells for high effective XXXXX', ['detector configuration', 'quantum efficiency'])\n", "('method used for task', 'ensemble XXXXX and rrelief algorithm were applied to establish the XXXXX', ['machine learning', 'quantitative assessment model'])\n", "('NONE', 'we use inverse probability of treatment weighting , a XXXXX based technique , for covariate adjustment of the cumulative incidence functions in competing XXXXX', ['propensity score', 'risk analysis'])\n", "('NONE', 'we use inverse probability of treatment weighting , a XXXXX based technique , for covariate adjustment of the XXXXX in competing risk analysis', ['propensity score', 'cumulative incidence functions'])\n", "('NONE', 'we use inverse probability of treatment weighting , a propensity score based technique , for covariate adjustment of the XXXXX in competing XXXXX', ['risk analysis', 'cumulative incidence functions'])\n", "('method used for task', 'XXXXX ( pca ) for dominant XXXXX', ['principal component analysis', 'feature selection'])\n", "('NONE', 'the XXXXX presented allows the deployment of different benchmarking tests for XXXXX', ['software tool', 'm2m protocols'])\n", "('NONE', 'the most relevant XXXXX were evaluated considering different specific XXXXX', ['performance metrics', 'm2m protocols'])\n", "('method used for task', 'the peristaltic flow of a copper oxide water fluid investigate the effects of XXXXX and XXXXX', ['heat generation', 'magnetic field'])\n", "('NONE', 'we have developed a freely available XXXXX for semi-automated tracking of muscle fascicles in b-mode XXXXX', ['software package', 'ultrasound image sequences'])\n", "('method used for task', 'XXXXX and a XXXXX of the lumbar spine allow the prediction of fracture probability', ['evolutionary algorithm', 'finite element model'])\n", "('NONE', 'XXXXX and a finite element model of the XXXXX allow the prediction of fracture probability', ['evolutionary algorithm', 'lumbar spine'])\n", "('NONE', 'XXXXX and a finite element model of the lumbar spine allow the prediction of XXXXX', ['evolutionary algorithm', 'fracture probability'])\n", "('NONE', 'evolutionary algorithm and a XXXXX of the XXXXX allow the prediction of fracture probability', ['finite element model', 'lumbar spine'])\n", "('NONE', 'evolutionary algorithm and a XXXXX of the lumbar spine allow the prediction of XXXXX', ['finite element model', 'fracture probability'])\n", "('NONE', 'evolutionary algorithm and a finite element model of the XXXXX allow the prediction of XXXXX', ['lumbar spine', 'fracture probability'])\n", "('NONE', 'one of the interesting parts of the study is the XXXXX that significantly affects XXXXX', ['parameter optimization', 'system performance'])\n", "('NONE', 'we have developed a XXXXX and an algorithm that allow XXXXX of abnormal values of vital parameters occurring during anesthesia', ['data model', 'automatic detection'])\n", "('NONE', 'we propose an XXXXX combining semi XXXXX ( ssl ) and active learning ( al ) for segmenting crohns disease affected regions in mri', ['interactive method', 'supervised learning'])\n", "('method used for task', 'we propose an XXXXX combining semi supervised learning ( ssl ) and XXXXX ( al ) for segmenting crohns disease affected regions in mri', ['interactive method', 'active learning'])\n", "('method used for task', 'we propose an interactive method combining semi XXXXX ( ssl ) and XXXXX ( al ) for segmenting crohns disease affected regions in mri', ['supervised learning', 'active learning'])\n", "('NONE', 'compared to fully supervised methods we obtain high XXXXX with fewer samples and lesser XXXXX', ['computation time', 'segmentation accuracy'])\n", "('NONE', 'we have created a software poped lite in order to increase the use of XXXXX in preclinical XXXXX', ['optimal design', 'drug discovery'])\n", "('method used for task', 'the proposed approach uses XXXXX to XXXXX', ['genetic algorithms', 'dimensionality reduction'])\n", "('method used for task', 'XXXXX and XXXXX are used to recognize the signals', ['k-nearest neighbor', 'naive bayes classifiers'])\n", "('NONE', 'the system achieved a high XXXXX and can be implemented in an XXXXX', ['classification accuracy', 'embedded system'])\n", "('NONE', 'the feature extraction method combines XXXXX , XXXXX and fisher distance', ['wavelet packet decomposition', 'common spatial patterns'])\n", "('NONE', 'the XXXXX combines XXXXX , common spatial patterns and fisher distance', ['wavelet packet decomposition', 'feature extraction method'])\n", "('NONE', 'the XXXXX combines wavelet packet decomposition , XXXXX and fisher distance', ['common spatial patterns', 'feature extraction method'])\n", "('NONE', 'an up-to-date review about the proposed techniques for the XXXXX of pigmented XXXXX is presented', ['image segmentation', 'skin lesions'])\n", "('NONE', 'investigating application of cnt in improving XXXXX of XXXXX', ['dynamic response', 'conical shell'])\n", "('method used for task', 'a rotation-based isogeometric reissner–mindlin shell formulation which is able to handle XXXXX and XXXXX is presented', ['finite rotations', 'large deformations'])\n", "('method used for task', 'a XXXXX phase-field approach to 3d XXXXX', ['higher order', 'brittle fracture'])\n", "('NONE', 'domain with singularities and/or XXXXX discretised into XXXXX', ['complex geometry', 'finite elements'])\n", "('NONE', 'the dual form of XXXXX , abbreviated as dda-d , with the XXXXX as the basic variables , is established', ['discontinuous deformation analysis', 'contact forces'])\n", "('method used for task', 'bézier extraction of truncated XXXXX and the application of the approach to adaptive XXXXX are presented', ['hierarchical b-splines', 'isogeometric analysis'])\n", "('NONE', 'a strict element viewpoint is adopted in the XXXXX which facilitates the application of standard procedures of XXXXX', ['refinement algorithms', 'adaptive finite element analysis'])\n", "('NONE', 'XXXXX demonstrate optimal XXXXX for problems that involve singularities and strong gradients', ['numerical examples', 'convergence rates'])\n", "('NONE', 'first XXXXX of the fractional XXXXX', ['numerical solution', 'navier–stokes equations'])\n", "('method used for task', 'stress XXXXX for contacting cracks are validated against XXXXX', ['analytical solutions', 'intensity factors'])\n", "('NONE', 'maximum entropy XXXXX handle dispersion errors better than XXXXX', ['basis functions', 'finite elements'])\n", "('NONE', 'explicit integral form of the XXXXX facilitates XXXXX', ['sensitivity analysis', 'distance constraint'])\n", "('method used for task', 'a subjective XXXXX on the XXXXX of http adaptive streaming ( has ) is conducted', ['user study', 'influence factors'])\n", "('method used for task', 'we model a cross-layer problem of XXXXX , link delay , and XXXXX', ['congestion control', 'power control'])\n", "('NONE', 'we simplify the auction procedure by allowing the users to skip the XXXXX when the user is unsure or unaware of the exact XXXXX of his own repeated jobs', ['utility function', 'utility model'])\n", "('NONE', 'we implement abacus by modifying the XXXXX in hadoop , and test it on a large-scale cloud platform . our experimental results verify the truthfulness of our auction-based mechanism , XXXXX , as well as the accuracy of our utility prediction algorithm', ['scheduling algorithm', 'system efficiency'])\n", "('NONE', 'we support the selection of a XXXXX depending on XXXXX', ['user requirements', 'modeling technique'])\n", "('method used for task', 'we study the possibility to employ XXXXX ( mt ) systems and supervised methods for multilingual XXXXX', ['machine translation', 'sentiment analysis'])\n", "('NONE', 'unlike traditional hybrid XXXXX , ours consists of two XXXXX', ['language models', 'independent components'])\n", "('NONE', 'global prosodic and laughter-specific XXXXX were extracted for the two types of laughter . these parameters were analysed by XXXXX and classification trees to reduce the number of parameters', ['acoustic features', 'principal component analysis'])\n", "('method used for task', 'a XXXXX was trained and tested using seven important features , and total XXXXX was confirmed to be at least over 84 with unseen test material', ['support vector machine', 'classification accuracy'])\n", "('method used for task', 'XXXXX is affected by environmental and interlocutor-related XXXXX', ['speech production', 'contextual factors'])\n", "('NONE', 'the class-specific XXXXX scheme improves the XXXXX', ['multiple classifiers', 'recognition accuracy'])\n", "('NONE', 'XXXXX are usually implemented XXXXX and difficult to adapt to new domains', ['ad hoc', 'dialog managers'])\n", "('method used for task', 'XXXXX for spoken XXXXX ( std ) on english ( meeting domain ) and spanish ( read speech ) data in a discriminative confidence estimation framework', ['feature analysis', 'term detection'])\n", "('method used for task', 'XXXXX is based on groups that are defined according to their XXXXX : lattice-based features , duration-based features , lexical features , levenshtein distance-based features , position and prosodic features ( pitch and energy )', ['feature analysis', 'information sources'])\n", "('method used for task', 'XXXXX employs two well-known and established models : XXXXX ( a generative approach ) and logistic XXXXX ( a discriminative approach ) . individual and incremental analyses are presented for both models', ['feature analysis', 'linear regression'])\n", "('method used for task', 'XXXXX employs two well-known and established models : XXXXX ( a generative approach ) and logistic XXXXX ( a discriminative approach ) . individual and incremental analyses are presented for both models', ['feature analysis', 'linear regression'])\n", "('method used for task', 'the best XXXXX comprises features from different groups : lattice-based and lexical features are among the most informative groups in general , and duration and energy are more informative for read XXXXX', ['feature set', 'speech data'])\n", "('method used for task', 'XXXXX for a spanish-to-lse XXXXX for wide-domain application', ['rule-based approach', 'machine translation system'])\n", "('NONE', 'application of e-pca for automatic XXXXX in a XXXXX', ['spoken dialogue system', 'belief compression'])\n", "('method used for task', 'first real XXXXX comparing manual and automatic XXXXX', ['user evaluation', 'belief compression'])\n", "('NONE', 'acoustic XXXXX between spoken segments can improve spoken XXXXX', ['feature similarity', 'term detection'])\n", "('NONE', 'detailed XXXXX of results using multiple XXXXX', ['comparative analysis', 'evaluation metrics'])\n", "('NONE', 'we model the XXXXX by a speaker adapted XXXXX and a vocal tract filter', ['speech signal', 'glottal source'])\n", "('method used for task', 'introduce novel XXXXX for XXXXX', ['emotion recognition', 'ranking models'])\n", "('method used for task', 'simplified and supervised i-vector modeling for robust and efficient XXXXX and XXXXX', ['language identification', 'speaker verification'])\n", "('NONE', 'lexical feature with XXXXX using XXXXX between words and dimension as weights is powerful', ['sparse representation', 'mutual information'])\n", "('NONE', 'XXXXX with both XXXXX and channel distortion', ['speech enhancement', 'additive noise'])\n", "('method used for task', 'use of XXXXX to reduce training and testing data mismatch for XXXXX', ['speech recognition', 'corpus data'])\n", "('method used for task', 'improved results on aurora 4 for XXXXX and XXXXX', ['speech recognition', 'speech enhancement'])\n", "('method used for task', 'XXXXX and XXXXX are used for proximity computation', ['euclidean distance', 'cosine dissimilarity'])\n", "('NONE', 'XXXXX unlike XXXXX makes weighted pagerank to converge', ['cosine similarity', 'cosine dissimilarity'])\n", "('NONE', 'refinements based on XXXXX further reduce XXXXX', ['predictive models', 'segmentation errors'])\n", "('method used for task', 'we explain in detail the different steps in computing a XXXXX based on a XXXXX', ['language model', 'recurrent neural network'])\n", "('method used for task', 'word-level XXXXX for perplexity-based XXXXX', ['linguistic information', 'data selection'])\n", "('method used for task', 'for that we perform an XXXXX automatic post-editing from ready-to-use generic XXXXX', ['online learning', 'machine translation systems'])\n", "('method used for task', 'a XXXXX is developed for determining the XXXXX', ['model parameters', 'parameter estimation method'])\n", "('NONE', 'a new type of pole-zero XXXXX of the XXXXX is obtained', ['transfer function', 'vocal tract'])\n", "('NONE', 'XXXXX in a hri domain shows that the approach outperforms traditional hand-crafted and XXXXX', ['user evaluation', 'statistical models'])\n", "('NONE', 'XXXXX with XXXXX that represent similarity measures', ['distance functions', 'probabilistic semantics'])\n", "('NONE', 'XXXXX with probabilistic semantics that represent XXXXX', ['distance functions', 'similarity measures'])\n", "('NONE', 'distance functions with XXXXX that represent XXXXX', ['probabilistic semantics', 'similarity measures'])\n", "('NONE', 'the proposed XXXXX can estimate the direction of a signal in a XXXXX', ['noisy environment', 'doa methods'])\n", "('NONE', 'XXXXX per unit type based synthesis system generates high XXXXX', ['neural network', 'speech quality'])\n", "('NONE', 'integrating the XXXXX in XXXXX', ['principal component analysis', 'wavelet packet decomposition'])\n", "('method used for task', 'the purpose of this contribution is to review the state of the art in both areas , XXXXX and XXXXX', ['statistical methods', 'speech processing'])\n", "('NONE', 'borrowing theories from XXXXX , we propose a XXXXX of human cognition', ['cognitive psychology', 'computational model'])\n", "('NONE', 'borrowing theories from XXXXX , we propose a computational model of XXXXX', ['cognitive psychology', 'human cognition'])\n", "('NONE', 'borrowing theories from cognitive psychology , we propose a XXXXX of XXXXX', ['computational model', 'human cognition'])\n", "('method used for task', 'we verify the XXXXX and XXXXX with narrative text data', ['cognitive model', 'summarization method'])\n", "('NONE', 'preprocessed elderly XXXXX were tested with an android XXXXX', ['smart phone', 'voice signals'])\n", "('method used for task', 'XXXXX increased to 1.5 % by increasing the XXXXX', ['speech rate', 'speech recognition accuracy'])\n", "('method used for task', 'a new , particular XXXXX is proposed to improve XXXXX for automatic speech recognition', ['logistic regression model', 'confidence measures'])\n", "('method used for task', 'a new , particular XXXXX is proposed to improve confidence measures for XXXXX', ['logistic regression model', 'automatic speech recognition'])\n", "('method used for task', 'a new , particular logistic regression model is proposed to improve XXXXX for XXXXX', ['confidence measures', 'automatic speech recognition'])\n", "('method used for task', 'we develop the XXXXX based on XXXXX', ['voice activity detection', 'statistical model'])\n", "('NONE', 'XXXXX can be defined by pos tags and automatic XXXXX', ['word clustering', 'word classes'])\n", "('method used for task', 'XXXXX are used to filter synthetic utterances for one XXXXX', ['acoustic features', 'input concept'])\n", "('method used for task', 'this work presents a novel comprehensive study on the use of XXXXX ( dnns ) for automatic XXXXX ( lid )', ['deep neural networks', 'language identification'])\n", "('method used for task', 'best XXXXX is achieved by combining the results from the proposed bottleneck system and the baseline XXXXX . compared to a state-of-the-art XXXXX , the combined system achieves a 45 % of relative improvement in both eer and cavg , on 3s and 10s', ['system performance', 'i-vector system'])\n", "('NONE', 'best system performance is achieved by combining the results from the proposed bottleneck system and the baseline XXXXX . compared to a state-of-the-art XXXXX , the combined system achieves a 45 % of relative improvement in both eer and cavg , on 3s and 10s', ['baseline system', 'i-vector system'])\n", "('method used for task', 'this special issue is devoted to XXXXX and XXXXX', ['natural language processing', 'human–computer interaction'])\n", "('NONE', 'originality of the topic for XXXXX in XXXXX', ['knowledge improvement', 'pdca methodology'])\n", "('NONE', 'the proposed pim is proven feasible for XXXXX via several XXXXX', ['field studies', 'soap frameworks'])\n", "('method used for task', 'the design and implementation of jujube XXXXX and XXXXX', ['real-time monitoring', 'management system'])\n", "('NONE', 'XXXXX improve the XXXXX significantly in sparse networks', ['unidirectional links', 'network lifetime'])\n", "('NONE', 'reference framework for XXXXX and improvement in XXXXX', ['process assessment', 'systems engineering'])\n", "('method used for task', 'XXXXX is employed to extract customer utilities for generating XXXXX in a customer-driven approach', ['conjoint analysis', 'design concepts'])\n", "('method used for task', 'med-trace is a lightweight XXXXX and XXXXX', ['traceability assessment', 'improvement method'])\n", "('NONE', 'based on the XXXXX , the XXXXX is proposed', ['routing algorithm', 'address configuration algorithm'])\n", "('NONE', 'an integrated platform of XXXXX in an XXXXX is proposed', ['data exchange', 'intelligent transportation systems'])\n", "('method used for task', 'most apps are developed without applying XXXXX and XXXXX', ['information security', 'best practices'])\n", "('method used for task', 'XXXXX for XXXXX', ['situational awareness', 'critical infrastructure protection'])\n", "('method used for task', 'a generic XXXXX for XXXXX data', ['data model', 'food consumption'])\n", "('method used for task', 'we have used XXXXX to define XXXXX', ['similarity measures', 'similarity networks'])\n", "('method used for task', 'we have defined a XXXXX based on XXXXX in order to establish trust along a path of entities', ['trust model', 'similarity networks'])\n", "('NONE', 'a XXXXX evaluates the reduction of bus utilization and initialization times under XXXXX', ['simulation study', 'data compression'])\n", "('method used for task', 'this paper proposes a XXXXX for developing solutions for XXXXX', ['software architecture', 'smart environments'])\n", "('NONE', 'XXXXX involves two direct benefits : less time of developing , and XXXXX', ['software architecture', 'cost reduction'])\n", "('method used for task', 'the XXXXX based on XXXXX is proposed', ['network architecture', 'location information'])\n", "('method used for task', 'XXXXX and XXXXX based method are proposed', ['image histogram', 'human vision system'])\n", "('NONE', 'XXXXX should be measured taking into account XXXXX', ['color images', 'human vision system'])\n", "('method used for task', 'context-aware XXXXX for XXXXX planning ( erp )', ['classification methodology', 'enterprise resource'])\n", "('NONE', 'the XXXXX can also be applied in XXXXX', ['dynamic environment', 'discovery scheme'])\n", "('method used for task', 'modelling the XXXXX requires a way to convey the XXXXX', ['dynamic constraints', 'state transitions'])\n", "('method used for task', 'we define a XXXXX for XXXXX by extending square XXXXX', ['quality model', 'semantic technologies'])\n", "('method used for task', 'we define a XXXXX for XXXXX by extending square XXXXX', ['semantic technologies', 'quality model'])\n", "('NONE', 'this research is in the field of hardware based XXXXX . even though XXXXX are the base of this work but the algorithm has many other applications', ['image processing', 'microarray images'])\n", "('NONE', 'hardware compression of XXXXX is just an example of XXXXX', ['microarray images', 'big data applications'])\n", "('NONE', 'field programmable XXXXX based XXXXX module reduces conversion time', ['gate array', 'analog input'])\n", "('NONE', 'field programmable XXXXX based analog input module reduces XXXXX', ['gate array', 'conversion time'])\n", "('NONE', 'field programmable gate array based XXXXX module reduces XXXXX', ['analog input', 'conversion time'])\n", "('NONE', 'an lp framework that is capable of jointly modeling XXXXX and route diversity in XXXXX is developed', ['energy dissipation', 'wireless sensor networks'])\n", "('NONE', 'an XXXXX that is capable of jointly modeling XXXXX and route diversity in wireless sensor networks is developed', ['energy dissipation', 'lp framework'])\n", "('method used for task', 'an lp framework that is capable of jointly modeling XXXXX and XXXXX in wireless sensor networks is developed', ['energy dissipation', 'route diversity'])\n", "('NONE', 'an XXXXX that is capable of jointly modeling energy dissipation and route diversity in XXXXX is developed', ['wireless sensor networks', 'lp framework'])\n", "('NONE', 'an lp framework that is capable of jointly modeling energy dissipation and XXXXX in XXXXX is developed', ['wireless sensor networks', 'route diversity'])\n", "('method used for task', 'an XXXXX that is capable of jointly modeling energy dissipation and XXXXX in wireless sensor networks is developed', ['lp framework', 'route diversity'])\n", "('method used for task', 'lp problems are solvable in XXXXX . hence , from computational point of view it is preferable to model a problem using lp rather than with a nonpolynomial time formulation which makes a comprehensive analysis infeasible . thus , the novel XXXXX presented in this paper can be used with minor modifications for future analysis on different aspects of route diversity', ['polynomial time', 'lp framework'])\n", "('method used for task', 'lp problems are solvable in XXXXX . hence , from computational point of view it is preferable to model a problem using lp rather than with a nonpolynomial time formulation which makes a comprehensive analysis infeasible . thus , the novel lp framework presented in this paper can be used with minor modifications for future analysis on different aspects of XXXXX', ['polynomial time', 'route diversity'])\n", "('NONE', 'lp problems are solvable in polynomial time . hence , from computational point of view it is preferable to model a problem using lp rather than with a nonpolynomial time formulation which makes a comprehensive analysis infeasible . thus , the novel XXXXX presented in this paper can be used with minor modifications for future analysis on different aspects of XXXXX', ['lp framework', 'route diversity'])\n", "('NONE', 'we propose schemes to ensure prioritization and reduce XXXXX in XXXXX', ['end-to-end delay', 'hybrid network'])\n", "('NONE', 'use the XXXXX generated by the XXXXX to generate a videogame', ['process models', 'domain experts'])\n", "('method used for task', 'XXXXX based on tripartite credibility for XXXXX are proposed', ['rfid systems', 'secure mechanisms'])\n", "('NONE', 'the XXXXX have advantages in communication and XXXXX', ['computational cost', 'secure mechanisms'])\n", "('method used for task', 'a new convolutional XXXXX with three convolutional layers and three fully-connected layers is introduced . XXXXX is utilized to model the output error of previous networks and adjust configurations of networks adaptively', ['network structure', 'gaussian distribution'])\n", "('NONE', 'we present the XXXXX of a focused XXXXX that combines link-based and content-based approaches to predict the topical focus of an unvisited page', ['architectural design', 'web crawler'])\n", "('NONE', 'we present a custom method using the dewey decimal XXXXX to best classify the subject of an unvisited page into standard human XXXXX', ['classification system', 'knowledge categories'])\n", "('NONE', 'to prioritize an unvisited url , we use a dynamic , flexible and updating XXXXX called t-graph . it helps find the XXXXX to get to on-topic pages on the web', ['hierarchical data structure', 'shortest path'])\n", "('NONE', 'the functional and XXXXX of a focused XXXXX are elicited and described , as well as standard evaluation criteria', ['non-functional requirements', 'web crawler'])\n", "('NONE', 'the functional and XXXXX of a focused web crawler are elicited and described , as well as standard XXXXX', ['non-functional requirements', 'evaluation criteria'])\n", "('NONE', 'the functional and non-functional requirements of a focused XXXXX are elicited and described , as well as standard XXXXX', ['web crawler', 'evaluation criteria'])\n", "('NONE', 'k-anycast communication technique is utilized for enhancing transmission reliability , load-balancing and security purpose by collecting multiple copies of a packet from a source and verifying the information of monitoring field . in r-wsns , due to energy replenished continually and limited energy storage capacity , a sensor can not be always beneficial to conserve energy when a network can harvest excessive energy from the environment . therefore , the surplus energy of sensor can be used for strengthening XXXXX . in this paper , a XXXXX for k-anycast communication based upon the anycast tree scheme is proposed for wireless sensor networks', ['data transmission', 'routing protocol'])\n", "('method used for task', 'k-anycast communication technique is utilized for enhancing transmission reliability , load-balancing and security purpose by collecting multiple copies of a packet from a source and verifying the information of monitoring field . in r-wsns , due to energy replenished continually and limited energy storage capacity , a sensor can not be always beneficial to conserve energy when a network can harvest excessive energy from the environment . therefore , the surplus energy of sensor can be used for strengthening XXXXX . in this paper , a routing protocol for k-anycast communication based upon the anycast tree scheme is proposed for XXXXX', ['data transmission', 'wireless sensor networks'])\n", "('method used for task', 'k-anycast communication technique is utilized for enhancing transmission reliability , load-balancing and security purpose by collecting multiple copies of a packet from a source and verifying the information of monitoring field . in r-wsns , due to energy replenished continually and limited energy storage capacity , a sensor can not be always beneficial to conserve energy when a network can harvest excessive energy from the environment . therefore , the surplus energy of sensor can be used for strengthening data transmission . in this paper , a XXXXX for k-anycast communication based upon the anycast tree scheme is proposed for XXXXX', ['routing protocol', 'wireless sensor networks'])\n", "('method used for task', 'multiple-metrics are utilized for instructing the XXXXX . a source initiates to create a XXXXX reaching any one sink with source node as the root', ['route discovery', 'spanning tree'])\n", "('NONE', 'state the problem of XXXXX in the XXXXX', ['service composition problem', 'future iot'])\n", "('NONE', 'review the state of the art of XXXXX in the XXXXX', ['service composition', 'future iot'])\n", "('NONE', 'provide research challenges of XXXXX in the XXXXX', ['service composition', 'future iot'])\n", "('NONE', 'we address the XXXXX that arise when outsourcing XXXXX in the cloud as bpaas ( business process as a service )', ['business processes', 'security issues'])\n", "('method used for task', 'language is defined formally via XXXXX and XXXXX', ['denotational semantics', 'attribute grammars'])\n", "('NONE', 'the proposal functions of the sga in this paper are suitable to be a new standard interface because their meet the basic XXXXX of the XXXXX layer', ['security requirements', 'm2m service'])\n", "('method used for task', 'this paper investigates density , XXXXX and XXXXX numbers', ['reference point', 'position error'])\n", "('NONE', 'we present a custom method using dewey decimal XXXXX to best classify the subject of an unvisited page into standard human XXXXX', ['classification system', 'knowledge categories'])\n", "('NONE', 'to prioritize an unvisited url , we use a dynamic , flexible and updating XXXXX called t-graph , which helps find the XXXXX to get to on-topic pages on the web', ['hierarchical data structure', 'shortest path'])\n", "('NONE', 'we present a uml-based domain specific XXXXX ( dsml ) that models the sa forum availability management framework ( amf ) XXXXX', ['modeling language', 'domain model'])\n", "('NONE', 'proposal of integration of iec 61850 scl with opc ua XXXXX in a XXXXX', ['information model', 'smart grid environment'])\n", "('method used for task', 'XXXXX and XXXXX of wearable devices are addressed ,', ['product differentiation', 'product selection'])\n", "('method used for task', 'arm ( XXXXX ) is adopted to identify attractive XXXXX ,', ['association rule mining', 'design attributes'])\n", "('NONE', 'shows a mashup platform that lets XXXXX make their own XXXXX', ['end users', 'web applications'])\n", "('method used for task', 'the XXXXX was conducted for three XXXXX , ehr submissions , queries , and retrievals , based on the ihe xds.b profile for ehr sharing', ['performance testing', 'use cases'])\n", "('NONE', 'proposing the XXXXX of cdnfre tool for detecting conflicts in XXXXX', ['software architecture', 'non-functional requirements'])\n", "('method used for task', 'XXXXX to improve the XXXXX', ['expert system', 'web accessibility'])\n", "('NONE', 'XXXXX among XXXXX , regions , and images are explored', ['hierarchical structure', 'feature points'])\n", "('NONE', 'we add a XXXXX to the process of XXXXX', ['hierarchical structure', 'sparse coding'])\n", "('NONE', 'the method learns a XXXXX of geometric transformation from the XXXXX set', ['shape space', '3d image'])\n", "('method used for task', 'XXXXX is registration between planning time and XXXXX in radiotherapy', ['target problem', 'treatment time'])\n", "('NONE', 'registration is performed on XXXXX of 17 patients with XXXXX', ['mr images', 'cervical cancer'])\n", "('NONE', 'XXXXX is developed in a variational framework using XXXXX', ['segmentation algorithm', 'level sets'])\n", "('NONE', 'this framework is based on XXXXX of 2d closed , XXXXX', ['shape analysis', 'planar curves'])\n", "('method used for task', 'the models can be used in XXXXX and 3d XXXXX', ['image registration', 'surface reconstruction'])\n", "('method used for task', 'the method is not overly sensitive with respect to XXXXX , missing frames and XXXXX', ['missing data', 'measurement noise'])\n", "('NONE', 'incorporate the XXXXX into an optimized search based XXXXX', ['shape model', 'segmentation method'])\n", "('method used for task', 'a XXXXX based on XXXXX is proposed', ['video segmentation', 'dynamic texture'])\n", "('method used for task', 'comparing with XXXXX , higher chance to find the XXXXX is achieved', ['chan–vese model', 'global solution'])\n", "('method used for task', 'we incorporate the spatial context to the conventional XXXXX ( fcm ) algorithm for XXXXX', ['fuzzy c-means', 'image segmentation'])\n", "('NONE', 'we propose a method for XXXXX from single XXXXX', ['pose estimation', 'ultrasound image'])\n", "('method used for task', 'a XXXXX correspondence algorithm is proposed for visual–thermal close-range XXXXX', ['video surveillance', 'stereo dense'])\n", "('NONE', 'such elastic XXXXX serve as XXXXX in a bayesian active contour model', ['shape models', 'shape priors'])\n", "('NONE', 'such elastic XXXXX serve as shape priors in a bayesian XXXXX', ['shape models', 'active contour model'])\n", "('NONE', 'such elastic shape models serve as XXXXX in a bayesian XXXXX', ['shape priors', 'active contour model'])\n", "('method used for task', 'we use a global XXXXX to segment leaves in complex XXXXX', ['shape model', 'natural images'])\n", "('method used for task', 'XXXXX and XXXXX improve segmentation performance', ['prior knowledge', 'shape constraints'])\n", "('method used for task', 'XXXXX and shape constraints improve XXXXX', ['prior knowledge', 'segmentation performance'])\n", "('NONE', 'prior knowledge and XXXXX improve XXXXX', ['shape constraints', 'segmentation performance'])\n", "('NONE', 'we present a comprehensive survey of XXXXX ( mrfs ) in XXXXX', ['markov random fields', 'computer vision'])\n", "('NONE', 'local memorization guides to search for the XXXXX that avoids XXXXX', ['optimal solution', 'local minima'])\n", "('method used for task', 'we describe and illustrate an XXXXX , based on the developed aggm , for XXXXX using infrared images', ['online algorithm', 'pedestrian detection'])\n", "('NONE', 'pca is the best XXXXX in very XXXXX less than 16*16pixels', ['recognition rate', 'low resolution'])\n", "('method used for task', 'it fits ellipses to XXXXX to analyse full XXXXX and motion', ['body parts', 'body shape'])\n", "('method used for task', 'opn could be utilized as XXXXX with semantics for scalable XXXXX', ['visual words', 'image retrieval'])\n", "('method used for task', 'extract XXXXX by an effective paradigm for XXXXX', ['semantic information', 'saliency detection'])\n", "('method used for task', 'a XXXXX based manifold representation for XXXXX', ['local feature', 'object tracking'])\n", "('method used for task', 'experimental results on both the synthetic XXXXX and recent ir database demonstrate jvim’s advantages over several recent XXXXX', ['shape models', 'cad data'])\n", "('method used for task', 'the data is calculated using gis-based XXXXX and XXXXX', ['augmented reality', 'computer vision'])\n", "('NONE', 'we use random walks to perform XXXXX with XXXXX', ['image segmentation', 'directed hypergraphs'])\n", "('method used for task', 'we propose a XXXXX for solving the XXXXX model', ['hybrid method', 'mumford–shah segmentation'])\n", "('method used for task', 'we show that the XXXXX is efficient for a wide range of XXXXX', ['hybrid method', 'model parameters'])\n", "('NONE', 'our results are close to XXXXX in XXXXX', ['human performance', 'expression recognition'])\n", "('method used for task', 'we applied multidimensional XXXXX , the matching metric , and a XXXXX to classify hepatic lesions', ['persistent homology', 'support vector machine'])\n", "('method used for task', 'application of XXXXX to the evaluation of a method for XXXXX', ['persistent homology', '3d reconstruction'])\n", "('NONE', 'reduce the length of the XXXXX using XXXXX', ['time series', 'dimensionality reduction techniques'])\n", "('method used for task', 'XXXXX for XXXXX', ['unsupervised learning', 'feature selection'])\n", "('NONE', 'our masks are based on XXXXX using a linear XXXXX', ['feature selection', 'support vector machine'])\n", "('method used for task', 'a XXXXX is linearized by rank relaxation to reduce the XXXXX', ['bilinear model', 'time complexity'])\n", "('method used for task', 'a novel XXXXX for the XXXXX is incorporated', ['uncertainty measure', 'motion estimation'])\n", "('method used for task', 'presents novel neuroscience inspired information theoretic approach to XXXXX based on XXXXX', ['motion segmentation', 'mutual information'])\n", "('NONE', 'comparative XXXXX against competing XXXXX', ['performance evaluation', 'segmentation methods'])\n", "('method used for task', 'XXXXX on the XXXXX are derived from affine correspondences', ['linear constraints', 'fundamental matrix'])\n", "('NONE', 'a XXXXX of the XXXXX by intersection of conics is proposed', ['fundamental matrix', 'calculation method'])\n", "('NONE', 'glss is built on global and local XXXXX that help boosting the efficacy of XXXXX', ['data structures', 'feature selection'])\n", "('method used for task', 'fast knn XXXXX is well integrated with submodular XXXXX', ['graph construction', 'dictionary learning'])\n", "('NONE', 'we present an objective XXXXX of statistical XXXXX', ['performance analysis', 'edge detection'])\n", "('NONE', 'we show how XXXXX can outperform traditional XXXXX', ['statistical tests', 'edge detection methods'])\n", "('NONE', 'introduce XXXXX into the XXXXX of quantization', ['statistical analysis', 'out-of-sample extension'])\n", "('method used for task', 'a semi-supervised multi-graph XXXXX is proposed for XXXXX', ['image search', 'hashing method'])\n", "('NONE', 'rao-blackwellized XXXXX with XXXXX', ['particle filtering', 'gaussian mixture models'])\n", "('NONE', 'the core contribution is the XXXXX of the XXXXX', ['formal proof', 'bijection principle'])\n", "('method used for task', 'two XXXXX , one on manifolds for appearance learning , another for XXXXX', ['particle filters', 'object tracking'])\n", "('method used for task', 'XXXXX to prevent XXXXX when objects are likely occluded', ['occlusion handling', 'online learning'])\n", "('NONE', 'our XXXXX operates directly on XXXXX via the sampson distance', ['cost function', 'image points'])\n", "('NONE', 'it is validated in XXXXX , XXXXX and shape retrieval scenarios', ['object recognition', '3d reconstruction'])\n", "('NONE', 'XXXXX of two distinctive examplars of XXXXX', ['theoretical analysis', 'statistical shape models'])\n", "('NONE', 'extensive XXXXX of two distinctive examplars of XXXXX', ['experimental comparison', 'statistical shape models'])\n", "('method used for task', 'we propose a new slice representation model of XXXXX for XXXXX', ['range data', 'head pose estimation'])\n", "('method used for task', 'we present a XXXXX to XXXXX for freely moving cameras', ['geometric approach', 'background subtraction'])\n", "('NONE', 'XXXXX are detected in complex scenes with significant XXXXX', ['moving objects', 'camera motion'])\n", "('method used for task', 'to do so we learn 3d XXXXX between objects and different XXXXX', ['spatial relationships', 'body parts'])\n", "('method used for task', 'we developed new XXXXX based on the XXXXX , hts and htsn', ['shape descriptors', 'hough transform'])\n", "('method used for task', 'a new XXXXX for XXXXX is proposed', ['evaluation procedure', 'action localization'])\n", "('NONE', 'a single XXXXX integrates out XXXXX', ['performance measure', 'quality constraints'])\n", "('NONE', 'soft XXXXX estimated from XXXXX', ['upper bounds', 'experimental data'])\n", "('NONE', 'a novel approach to determine adaptively the XXXXX of XXXXX', ['temporal consistency', 'particle filters'])\n", "('method used for task', 'proposed method provides satisfactory XXXXX and XXXXX', ['detection accuracy', 'generalization capability'])\n", "('method used for task', 'indepth XXXXX and XXXXX', ['experimental validation', 'error analysis'])\n", "('NONE', 'dominant XXXXX by exploring XXXXX with histogram', ['gradient information', 'pixel selection'])\n", "('NONE', 'we encode pairwise relative XXXXX of patches in the bag of XXXXX', ['spatial information', 'word model'])\n", "('NONE', 'a new XXXXX optimized with XXXXX is proposed', ['objective function', 'graph cuts'])\n", "('method used for task', 'new and well-known XXXXX and XXXXX are both used', ['feature extraction', 'classification methods'])\n", "('method used for task', 'object matching and XXXXX is combined for achieving automatic XXXXX', ['active contour', 'object segmentation'])\n", "('method used for task', 'XXXXX based on likelihood and XXXXX without line search', ['learning algorithms', 'margin maximization'])\n", "('NONE', 'we formulate XXXXX as a statistical XXXXX', ['image segmentation', 'parameter estimation problem'])\n", "('method used for task', 'thorough XXXXX to compare the performances of several algorithms for hierarchical XXXXX', ['experimental study', 'image classification'])\n", "('NONE', 'we recover simultaneously the non-rigid XXXXX and the corresponding camera pose from a single image in XXXXX', ['3d shape', 'real time'])\n", "('NONE', 'high XXXXX in two computer vision applications : XXXXX and human action recognition', ['classification accuracy', 'facial expression recognition'])\n", "('NONE', 'high XXXXX in two computer vision applications : facial expression recognition and XXXXX', ['classification accuracy', 'human action recognition'])\n", "('method used for task', 'high classification accuracy in two computer vision applications : XXXXX and XXXXX', ['facial expression recognition', 'human action recognition'])\n", "('NONE', 'projection of the XXXXX from the XXXXX to the tangent euclidean space', ['covariance matrix', 'riemannian manifold'])\n", "('method used for task', 'projection of the XXXXX from the riemannian manifold to the tangent XXXXX', ['covariance matrix', 'euclidean space'])\n", "('method used for task', 'projection of the covariance matrix from the XXXXX to the tangent XXXXX', ['riemannian manifold', 'euclidean space'])\n", "('NONE', 'the algorithm is efficient both in terms of XXXXX and of XXXXX', ['computational cost', 'detection performance'])\n", "('NONE', 'based on XXXXX , a hybrid XXXXX is proposed', ['kernel regression', 'fusion strategy'])\n", "('method used for task', 'a XXXXX for each XXXXX', ['closed-form solution', 'camera model'])\n", "('method used for task', 'an XXXXX comparing the algebraic procedures to global polynomial optimization and an XXXXX', ['experimental evaluation', 'interior-point method'])\n", "('NONE', 'this method leverages discriminative computer vision models for faster XXXXX in XXXXX', ['probabilistic inference', 'generative models'])\n", "('NONE', 'a two-stage XXXXX system using XXXXX is proposed', ['head detection', 'motion features'])\n", "('method used for task', 'geometric normalization is critical to most XXXXX ( fr ) algorithms and is usually based off XXXXX', ['face recognition', 'eye locations'])\n", "('NONE', 'the XXXXX obtained from the proposed algorithm out perform other tested algorithms in both XXXXX and fr results', ['eye detection', 'eye locations'])\n", "('method used for task', 'a new 3d local XXXXX for XXXXX and human action recognition from depth sensors', ['shape descriptor', 'hand gesture'])\n", "('method used for task', 'a new 3d local XXXXX for hand gesture and XXXXX from depth sensors', ['shape descriptor', 'human action recognition'])\n", "('method used for task', 'a new 3d local shape descriptor for XXXXX and XXXXX from depth sensors', ['hand gesture', 'human action recognition'])\n", "('method used for task', 'XXXXX for the direct XXXXX based camera calibration', ['uncertainty analysis', 'linear transformation'])\n", "('method used for task', 'XXXXX for the direct linear transformation based XXXXX', ['uncertainty analysis', 'camera calibration'])\n", "('NONE', 'uncertainty analysis for the direct XXXXX based XXXXX', ['linear transformation', 'camera calibration'])\n", "('method used for task', 'we propose a method for XXXXX on 3d and 4d XXXXX', ['range data', 'landmark localization'])\n", "('NONE', 'we study object motion in stereo XXXXX by providing a novel XXXXX', ['mathematical analysis', 'video content'])\n", "('method used for task', 'the XXXXX is described by weighted XXXXX', ['object model', 'gaussian mixture models'])\n", "('NONE', 'sin parses structured XXXXX ( or XXXXX in general )', ['time series data', 'activity sequence'])\n", "('method used for task', 'hand-gesture XXXXX based on XXXXX for hci', ['recognition system', 'color imagery'])\n", "('NONE', 'a redundant XXXXX has been composed by using XXXXX', ['wavelet transform', 'texture feature space'])\n", "('NONE', 'the XXXXX method resembles dense surface shape recovery from XXXXX', ['missing data', '3d face estimation'])\n", "('method used for task', 'combining XXXXX and XXXXX with multiplication performs better for nearly planar objects', ['edge information', 'surface normals'])\n", "('method used for task', 'two well-known state-of-the-art methods and one XXXXX are extended to use this XXXXX', ['semantic information', 'baseline method'])\n", "('NONE', 'novel aam based XXXXX and combination of XXXXX and XXXXX', ['hand gesture', 'face features'])\n", "('NONE', 'novel aam based XXXXX and combination of XXXXX and XXXXX', ['hand gesture', 'face features'])\n", "('method used for task', 'we propose a hierarchical kernel based method to learn the non-linear correlations between rgb and depth XXXXX for XXXXX', ['action recognition', 'action data'])\n", "('method used for task', 'part models are formulated with a XXXXX for XXXXX', ['graph structure', 'object tracking'])\n", "('NONE', 'XXXXX are formulated with a XXXXX for object tracking', ['graph structure', 'part models'])\n", "('method used for task', 'XXXXX are formulated with a graph structure for XXXXX', ['object tracking', 'part models'])\n", "('method used for task', 'weighted XXXXX are used to handle XXXXX change and occlusion', ['part models', 'target appearance'])\n", "('method used for task', 'XXXXX are used to control the sample selections for XXXXX update', ['weight models', 'part model'])\n", "('method used for task', 'we experiment on a high variety of scenarios and public datasets ( genre classification - blip10000 , XXXXX - ucf50 / ucf101 and daily XXXXX - adl ) and show the benefits of the proposed approach which outperforms other state of the art approaches', ['action recognition', 'activities recognition'])\n", "('NONE', 'coupling detection and XXXXX of XXXXX in a single function', ['data association', 'object tracking'])\n", "('method used for task', 'XXXXX is incorporated with XXXXX by extracting the interactive contours', ['contextual information', 'motion features'])\n", "('NONE', 'a latent XXXXX integrating XXXXX , group discovery , and activity recognition is proposed', ['graphical model', 'multi-target tracking'])\n", "('method used for task', 'a latent XXXXX integrating multi-target tracking , group discovery , and XXXXX is proposed', ['graphical model', 'activity recognition'])\n", "('method used for task', 'a latent graphical model integrating XXXXX , group discovery , and XXXXX is proposed', ['multi-target tracking', 'activity recognition'])\n", "('NONE', 'performance of XXXXX improves when XXXXX and group clustering are incorporated', ['activity recognition', 'multi-target tracking'])\n", "('method used for task', 'our method showed superior performance on both XXXXX and realistic dataset on action and XXXXX', ['synthetic data', 'gesture recognition'])\n", "('NONE', 'the framework allows the creation of a high-level XXXXX of the scene and scalable XXXXX', ['semantic representation', 'feature extraction'])\n", "('method used for task', 'an XXXXX for inference in binary XXXXX mrf–map is proposed', ['efficient algorithm', 'higher order'])\n", "('method used for task', 'a XXXXX is formulated as a XXXXX discovering problem', ['multi-target tracking', 'dense subgraph'])\n", "('NONE', 'XXXXX : XXXXX are clearly visible in different light conditions', ['robust solution', 'traffic lights'])\n", "('method used for task', 'XXXXX can provide semantic cue to enhance the XXXXX', ['object recognition', 'material recognition'])\n", "('NONE', 'dense subgraphs convey more information about local XXXXX than simple XXXXX', ['graph structure', 'centrality measures'])\n", "('NONE', 'XXXXX convey more information about local XXXXX than simple centrality measures', ['graph structure', 'dense subgraphs'])\n", "('NONE', 'XXXXX convey more information about local graph structure than simple XXXXX', ['centrality measures', 'dense subgraphs'])\n", "('method used for task', 'region compactness in addition to XXXXX and XXXXX has been used to compute image region similarities', ['color features', 'image intensity'])\n", "('method used for task', 'the framework yields a simple , efficient , XXXXX for XXXXX', ['closed-form solution', 'change detection'])\n", "('method used for task', 'we detach the intrinsic XXXXX related to either brightness or XXXXX', ['camera parameters', 'depth data'])\n", "('NONE', 'the adce algorithm can find the salient XXXXX of XXXXX', ['feature points', 'shape contour'])\n", "('method used for task', 'the paper proposes a geometric error based triangulation and metric study of lines . as one of the fundamental problems in XXXXX , line triangulation is to determine the 3d coordinates of a line based on its 2d image projections from more than two views of cameras . compared to point features , XXXXX are more robust to matching errors , occlusions , and image uncertainties . however , when the number of views is larger than two , the back-projection planes usually do not intersect at one line due to measurement errors and image noise . thus , it is critical to find a 3d line that optimally fits the measured data . this paper , at first time , provides a comprehensive study on the triangulation and metric of lines based on geometric error', ['computer vision', 'line segments'])\n", "('method used for task', 'the paper proposes a geometric error based triangulation and metric study of lines . as one of the fundamental problems in XXXXX , line triangulation is to determine the 3d coordinates of a line based on its 2d image projections from more than two views of cameras . compared to point features , line segments are more robust to matching errors , occlusions , and image uncertainties . however , when the number of views is larger than two , the back-projection planes usually do not intersect at one line due to XXXXX and image noise . thus , it is critical to find a 3d line that optimally fits the measured data . this paper , at first time , provides a comprehensive study on the triangulation and metric of lines based on geometric error', ['computer vision', 'measurement errors'])\n", "('NONE', 'the paper proposes a geometric error based triangulation and metric study of lines . as one of the fundamental problems in XXXXX , XXXXX is to determine the 3d coordinates of a line based on its 2d image projections from more than two views of cameras . compared to point features , line segments are more robust to matching errors , occlusions , and image uncertainties . however , when the number of views is larger than two , the back-projection planes usually do not intersect at one line due to measurement errors and image noise . thus , it is critical to find a 3d line that optimally fits the measured data . this paper , at first time , provides a comprehensive study on the triangulation and metric of lines based on geometric error', ['computer vision', 'line triangulation'])\n", "('method used for task', 'the paper proposes a geometric error based triangulation and metric study of lines . as one of the fundamental problems in computer vision , line triangulation is to determine the 3d coordinates of a line based on its 2d image projections from more than two views of cameras . compared to point features , XXXXX are more robust to matching errors , occlusions , and image uncertainties . however , when the number of views is larger than two , the back-projection planes usually do not intersect at one line due to XXXXX and image noise . thus , it is critical to find a 3d line that optimally fits the measured data . this paper , at first time , provides a comprehensive study on the triangulation and metric of lines based on geometric error', ['line segments', 'measurement errors'])\n", "('method used for task', 'the paper proposes a geometric error based triangulation and metric study of lines . as one of the fundamental problems in computer vision , XXXXX is to determine the 3d coordinates of a line based on its 2d image projections from more than two views of cameras . compared to point features , XXXXX are more robust to matching errors , occlusions , and image uncertainties . however , when the number of views is larger than two , the back-projection planes usually do not intersect at one line due to measurement errors and image noise . thus , it is critical to find a 3d line that optimally fits the measured data . this paper , at first time , provides a comprehensive study on the triangulation and metric of lines based on geometric error', ['line segments', 'line triangulation'])\n", "('method used for task', 'the paper proposes a geometric error based triangulation and metric study of lines . as one of the fundamental problems in computer vision , XXXXX is to determine the 3d coordinates of a line based on its 2d image projections from more than two views of cameras . compared to point features , line segments are more robust to matching errors , occlusions , and image uncertainties . however , when the number of views is larger than two , the back-projection planes usually do not intersect at one line due to XXXXX and image noise . thus , it is critical to find a 3d line that optimally fits the measured data . this paper , at first time , provides a comprehensive study on the triangulation and metric of lines based on geometric error', ['measurement errors', 'line triangulation'])\n", "('method used for task', 'in this paper , a comprehensive study of line triangulation is conducted using geometric XXXXX . compared to the algebraic error based approaches , geometric error based algorithm is more meaningful , and thus , yields better estimation results . the main contributions of this study include : ( i ) it is proved that the XXXXX to minimizing the geometric errors can be transformed to finding the real roots of algebraic equations ; ( ii ) an effective iterative algorithm , iteg , is proposed to minimizing the geometric errors ; and ( iii ) an in-depth comparative evaluations on three metrics in 3d line space , the euclidean metric , the orthogonal metric , and the quasi-riemannian metric , are carried out . extensive experiments on synthetic data and real images are carried out to validate and demonstrate the effectiveness of the proposed algorithms', ['cost functions', 'optimal solution'])\n", "('method used for task', 'in this paper , a comprehensive study of line triangulation is conducted using geometric XXXXX . compared to the algebraic error based approaches , geometric error based algorithm is more meaningful , and thus , yields better estimation results . the main contributions of this study include : ( i ) it is proved that the optimal solution to minimizing the geometric errors can be transformed to finding the real roots of algebraic equations ; ( ii ) an effective XXXXX , iteg , is proposed to minimizing the geometric errors ; and ( iii ) an in-depth comparative evaluations on three metrics in 3d line space , the euclidean metric , the orthogonal metric , and the quasi-riemannian metric , are carried out . extensive experiments on synthetic data and real images are carried out to validate and demonstrate the effectiveness of the proposed algorithms', ['cost functions', 'iterative algorithm'])\n", "('method used for task', 'in this paper , a comprehensive study of line triangulation is conducted using geometric XXXXX . compared to the algebraic error based approaches , geometric error based algorithm is more meaningful , and thus , yields better estimation results . the main contributions of this study include : ( i ) it is proved that the optimal solution to minimizing the geometric errors can be transformed to finding the real roots of algebraic equations ; ( ii ) an effective iterative algorithm , iteg , is proposed to minimizing the geometric errors ; and ( iii ) an in-depth comparative evaluations on three metrics in 3d line space , the euclidean metric , the orthogonal metric , and the quasi-riemannian metric , are carried out . extensive experiments on XXXXX and real images are carried out to validate and demonstrate the effectiveness of the proposed algorithms', ['cost functions', 'synthetic data'])\n", "('method used for task', 'in this paper , a comprehensive study of XXXXX is conducted using geometric XXXXX . compared to the algebraic error based approaches , geometric error based algorithm is more meaningful , and thus , yields better estimation results . the main contributions of this study include : ( i ) it is proved that the optimal solution to minimizing the geometric errors can be transformed to finding the real roots of algebraic equations ; ( ii ) an effective iterative algorithm , iteg , is proposed to minimizing the geometric errors ; and ( iii ) an in-depth comparative evaluations on three metrics in 3d line space , the euclidean metric , the orthogonal metric , and the quasi-riemannian metric , are carried out . extensive experiments on synthetic data and real images are carried out to validate and demonstrate the effectiveness of the proposed algorithms', ['cost functions', 'line triangulation'])\n", "('method used for task', 'in this paper , a comprehensive study of line triangulation is conducted using geometric cost functions . compared to the algebraic error based approaches , geometric error based algorithm is more meaningful , and thus , yields better estimation results . the main contributions of this study include : ( i ) it is proved that the XXXXX to minimizing the geometric errors can be transformed to finding the real roots of algebraic equations ; ( ii ) an effective XXXXX , iteg , is proposed to minimizing the geometric errors ; and ( iii ) an in-depth comparative evaluations on three metrics in 3d line space , the euclidean metric , the orthogonal metric , and the quasi-riemannian metric , are carried out . extensive experiments on synthetic data and real images are carried out to validate and demonstrate the effectiveness of the proposed algorithms', ['optimal solution', 'iterative algorithm'])\n", "('method used for task', 'in this paper , a comprehensive study of line triangulation is conducted using geometric cost functions . compared to the algebraic error based approaches , geometric error based algorithm is more meaningful , and thus , yields better estimation results . the main contributions of this study include : ( i ) it is proved that the XXXXX to minimizing the geometric errors can be transformed to finding the real roots of algebraic equations ; ( ii ) an effective iterative algorithm , iteg , is proposed to minimizing the geometric errors ; and ( iii ) an in-depth comparative evaluations on three metrics in 3d line space , the euclidean metric , the orthogonal metric , and the quasi-riemannian metric , are carried out . extensive experiments on XXXXX and real images are carried out to validate and demonstrate the effectiveness of the proposed algorithms', ['optimal solution', 'synthetic data'])\n", "('method used for task', 'in this paper , a comprehensive study of XXXXX is conducted using geometric cost functions . compared to the algebraic error based approaches , geometric error based algorithm is more meaningful , and thus , yields better estimation results . the main contributions of this study include : ( i ) it is proved that the XXXXX to minimizing the geometric errors can be transformed to finding the real roots of algebraic equations ; ( ii ) an effective iterative algorithm , iteg , is proposed to minimizing the geometric errors ; and ( iii ) an in-depth comparative evaluations on three metrics in 3d line space , the euclidean metric , the orthogonal metric , and the quasi-riemannian metric , are carried out . extensive experiments on synthetic data and real images are carried out to validate and demonstrate the effectiveness of the proposed algorithms', ['optimal solution', 'line triangulation'])\n", "('method used for task', 'in this paper , a comprehensive study of line triangulation is conducted using geometric cost functions . compared to the algebraic error based approaches , geometric error based algorithm is more meaningful , and thus , yields better estimation results . the main contributions of this study include : ( i ) it is proved that the optimal solution to minimizing the geometric errors can be transformed to finding the real roots of algebraic equations ; ( ii ) an effective XXXXX , iteg , is proposed to minimizing the geometric errors ; and ( iii ) an in-depth comparative evaluations on three metrics in 3d line space , the euclidean metric , the orthogonal metric , and the quasi-riemannian metric , are carried out . extensive experiments on XXXXX and real images are carried out to validate and demonstrate the effectiveness of the proposed algorithms', ['iterative algorithm', 'synthetic data'])\n", "('method used for task', 'in this paper , a comprehensive study of XXXXX is conducted using geometric cost functions . compared to the algebraic error based approaches , geometric error based algorithm is more meaningful , and thus , yields better estimation results . the main contributions of this study include : ( i ) it is proved that the optimal solution to minimizing the geometric errors can be transformed to finding the real roots of algebraic equations ; ( ii ) an effective XXXXX , iteg , is proposed to minimizing the geometric errors ; and ( iii ) an in-depth comparative evaluations on three metrics in 3d line space , the euclidean metric , the orthogonal metric , and the quasi-riemannian metric , are carried out . extensive experiments on synthetic data and real images are carried out to validate and demonstrate the effectiveness of the proposed algorithms', ['iterative algorithm', 'line triangulation'])\n", "('method used for task', 'in this paper , a comprehensive study of XXXXX is conducted using geometric cost functions . compared to the algebraic error based approaches , geometric error based algorithm is more meaningful , and thus , yields better estimation results . the main contributions of this study include : ( i ) it is proved that the optimal solution to minimizing the geometric errors can be transformed to finding the real roots of algebraic equations ; ( ii ) an effective iterative algorithm , iteg , is proposed to minimizing the geometric errors ; and ( iii ) an in-depth comparative evaluations on three metrics in 3d line space , the euclidean metric , the orthogonal metric , and the quasi-riemannian metric , are carried out . extensive experiments on XXXXX and real images are carried out to validate and demonstrate the effectiveness of the proposed algorithms', ['synthetic data', 'line triangulation'])\n", "('NONE', 'a system with dynamic lighting is proposed that predicts the validity of a photometric model without XXXXX of the XXXXX', ['prior knowledge', 'scene geometry'])\n", "('method used for task', 'a XXXXX to improve the XXXXX given the user’s position , available sensors , and recognition tools is presented', ['speech recognition', 'fusion method'])\n", "('NONE', 'gui for clinical uses allows real-time , XXXXX with minimal XXXXX', ['high performance', 'user feedback'])\n", "('method used for task', 'XXXXX are adopted for tooth brushing XXXXX', ['hidden markov models', 'gesture recognition'])\n", "('NONE', 'we develop a XXXXX as a type of XXXXX', ['assistive technology', 'hci system'])\n", "('method used for task', 'we introduce a new dataset , the video water database , for XXXXX and to encourage research into XXXXX', ['experimental evaluation', 'water detection'])\n", "('method used for task', 'we show experimentally that our water detection method improves over methods from XXXXX and XXXXX', ['dynamic texture', 'material recognition'])\n", "('NONE', 'we show experimentally that our XXXXX method improves over methods from XXXXX and material recognition', ['dynamic texture', 'water detection'])\n", "('method used for task', 'we show experimentally that our XXXXX method improves over methods from dynamic texture and XXXXX', ['material recognition', 'water detection'])\n", "('method used for task', 'online XXXXX and quality visualization for XXXXX', ['user feedback', 'image acquisition'])\n", "('method used for task', 'we revisit three tasks : image sensing , XXXXX and XXXXX', ['scene segmentation', 'optical flow'])\n", "('NONE', 'a methodology for detecting the XXXXX position in XXXXX is presented', ['fundus images', 'fovea center'])\n", "('method used for task', 'XXXXX : methodology-provided and actual XXXXX distance', ['evaluation criterion', 'fovea center'])\n", "('method used for task', 'advanced texture analysis methods – XXXXX and gaussian XXXXX', ['local binary patterns', 'markov random fields'])\n", "('NONE', 'early XXXXX with lsa outperforms late XXXXX', ['data fusion', 'fusion methods'])\n", "('NONE', 'low-level fusion of the XXXXX slightly improves the XXXXX', ['visual features', 'predictive performance'])\n", "('method used for task', 'as a result of improved XXXXX compared to previous methods , needle XXXXX and robustness are increased using the proposed method', ['segmentation performance', 'localization accuracy'])\n", "('method used for task', 'the area under XXXXX increased from 0.78 to 0.86 with the XXXXX', ['roc curve', 'data-driven approach'])\n", "('NONE', 'XXXXX that significantly reduces the XXXXX', ['efficient implementation', 'processing time'])\n", "('method used for task', 'a new two-layer structural XXXXX for detecting microscopic image cells , which focuses on the representation issue to propose a model to effectively capture the rich XXXXX', ['contextual information', 'prediction framework'])\n", "('method used for task', 'we presented an XXXXX that is tailored to model and analyze the two-channel images of XXXXX', ['image processing pipeline', 'muscle fibers'])\n", "('NONE', 'the method quantifies morphological and XXXXX of nuclei and cytoplasm of XXXXX and derives other measurements about XXXXX', ['geometric features', 'muscle fibers'])\n", "('NONE', 'the method quantifies morphological and XXXXX of nuclei and cytoplasm of XXXXX and derives other measurements about XXXXX', ['geometric features', 'muscle fibers'])\n", "('method used for task', 'a detailed review of automated XXXXX for the pulmonary lobes from clinical XXXXX', ['segmentation methods', 'ct data'])\n", "('method used for task', 'we proposed a XXXXX based method to detect the respiratory signal from 2d XXXXX', ['manifold learning', 'ultrasound images'])\n", "('NONE', 'we propose the voxel visibility model for XXXXX in XXXXX', ['transfer function', 'volume rendering'])\n", "('method used for task', 'we propose the XXXXX model for XXXXX in volume rendering', ['transfer function', 'voxel visibility'])\n", "('method used for task', 'we propose the XXXXX model for transfer function in XXXXX', ['volume rendering', 'voxel visibility'])\n", "('method used for task', 'we propose an XXXXX for automatic XXXXX', ['optimization algorithm', 'transfer function design'])\n", "('method used for task', 'memory based XXXXX that is robust to pixel-level XXXXX', ['active contour', 'classification errors'])\n", "('NONE', 'test 7180 crypts : 87 % and 9 % true and XXXXX , 96 % XXXXX', ['false positives', 'segmentation accuracy'])\n", "('method used for task', 'evaluations over a rv XXXXX and hlhs XXXXX show the robustness of the method', ['challenge data', 'patient data'])\n", "('NONE', 'we perform XXXXX of healthy and tumor areas in XXXXX of bcc skin samples', ['automated classification', 'cars images'])\n", "('NONE', 'we believe this is an important step towards automated XXXXX in XXXXX', ['tumor detection', 'cars images'])\n", "('NONE', 'implementation of XXXXX in a XXXXX organ extraction procedure', ['granular computing', 'model based'])\n", "('method used for task', 'a XXXXX is used for XXXXX and svm parameter selection', ['genetic algorithm', 'feature selection'])\n", "('NONE', 'XXXXX in XXXXX tasks', ['experimental validation', 'cocaine addiction'])\n", "('method used for task', 'building an XXXXX able to process the XXXXX', ['expert system', 'feature vector'])\n", "('method used for task', 'XXXXX and XXXXX are used during feature extraction', ['principal component analysis', 'radon transform'])\n", "('method used for task', 'XXXXX and radon transform are used during XXXXX', ['principal component analysis', 'feature extraction'])\n", "('method used for task', 'principal component analysis and XXXXX are used during XXXXX', ['radon transform', 'feature extraction'])\n", "('NONE', 'we analyzed the XXXXX after XXXXX', ['healing process', 'bone loss'])\n", "('NONE', 'XXXXX decreases during the XXXXX after bone loss', ['fractal dimension', 'healing process'])\n", "('NONE', 'XXXXX decreases during the healing process after XXXXX', ['fractal dimension', 'bone loss'])\n", "('NONE', 'fractal dimension decreases during the XXXXX after XXXXX', ['healing process', 'bone loss'])\n", "('method used for task', 'a fully automatic XXXXX is presented to detect XXXXX using mri', ['cad system', 'prostate cancer'])\n", "('NONE', 'XXXXX have an advantage over XXXXX', ['mr images', 'ct images'])\n", "('NONE', 'we improved XXXXX by adaptation of additional training data and interpolating XXXXX', ['language models', 'translation quality'])\n", "('method used for task', 'system aims to provide longitudinal tracking , XXXXX , and XXXXX', ['data mining', 'data analysis'])\n", "('NONE', 'choroidal thickness and XXXXX are given vis-à-vis XXXXX', ['volume estimates', 'observer repeatability'])\n", "('method used for task', 'XXXXX had little dependency on the time after XXXXX', ['quantitative imaging features', 'contrast injection'])\n", "('NONE', 'some XXXXX changed gradually after XXXXX but eventually became stable', ['texture features', 'contrast injection'])\n", "('method used for task', 'energy functional includes XXXXX for XXXXX and level set function', ['total variation regularization', 'bias field'])\n", "('method used for task', 'energy functional includes XXXXX for bias field and XXXXX', ['total variation regularization', 'level set function'])\n", "('method used for task', 'energy functional includes total variation regularization for XXXXX and XXXXX', ['bias field', 'level set function'])\n", "('method used for task', 'a multiscale XXXXX for XXXXX is proposed', ['region growing method', 'coronary artery segmentation'])\n", "('method used for task', 'the first multi-center milestone XXXXX for vertebra XXXXX', ['comparative study', 'segmentation methods'])\n", "('method used for task', 'a new energy aware XXXXX has been proposed for cluster based XXXXX', ['routing algorithm', 'wireless sensor networks'])\n", "('NONE', 'it achieves o ( 1 ) XXXXX per XXXXX and o ( n ) time complexity for a wsn having n sensor nodes', ['message complexity', 'sensor node'])\n", "('method used for task', 'it achieves o ( 1 ) XXXXX per sensor node and o ( n ) XXXXX for a wsn having n sensor nodes', ['message complexity', 'time complexity'])\n", "('method used for task', 'it achieves o ( 1 ) XXXXX per sensor node and o ( n ) time complexity for a wsn having n XXXXX', ['message complexity', 'sensor nodes'])\n", "('method used for task', 'it achieves o ( 1 ) message complexity per XXXXX and o ( n ) XXXXX for a wsn having n sensor nodes', ['sensor node', 'time complexity'])\n", "('method used for task', 'it achieves o ( 1 ) message complexity per XXXXX and o ( n ) time complexity for a wsn having n XXXXX', ['sensor node', 'sensor nodes'])\n", "('method used for task', 'it achieves o ( 1 ) message complexity per sensor node and o ( n ) XXXXX for a wsn having n XXXXX', ['time complexity', 'sensor nodes'])\n", "('method used for task', 'it is successful in balancing the relaying load among the XXXXX with respect to their XXXXX', ['residual energy', 'sensor nodes'])\n", "('method used for task', 'XXXXX is solved by XXXXX and compared the performance with pso and fa', ['differential evolution algorithm', 'sa problem'])\n", "('method used for task', 'face allows XXXXX to customize XXXXX , localization , and processing procedures', ['data partitioning', 'application developers'])\n", "('method used for task', 'establishing the security-aware task , security overhead and XXXXX for aperiodic XXXXX', ['real-time applications', 'risk models'])\n", "('NONE', 'we achieve the XXXXX without inviting large XXXXX', ['storage capacity', 'communication overhead'])\n", "('method used for task', 'an sdma-based XXXXX ( s-mac ) for XXXXX is proposed', ['mac protocol', 'wireless ad hoc networks'])\n", "('method used for task', 'the proposed method is novel as it is important to install low cost sensors and XXXXX along with induction machines to achieve short XXXXX and an automated way of reporting the fault', ['detection mechanisms', 'detection time'])\n", "('NONE', 'a novel XXXXX 12t mtcmos based XXXXX is proposed', ['low power', 'sram cell'])\n", "('method used for task', 'XXXXX reduces the effect on XXXXX caused by illumination', ['histogram equalization', 'msa features'])\n", "('method used for task', 'the number of ( α , β ) pairs affects the size of XXXXX and XXXXX', ['msa features', 'recognition rates'])\n", "('method used for task', 'we propose semantic-based XXXXX through deductive logic and XXXXX', ['policy analysis', 'inference rules'])\n", "('method used for task', 'we present flaw , conflict and XXXXX algorithms for XXXXX', ['redundancy detection', 'xacml policy analysis'])\n", "('NONE', 'an architectural and XXXXX of both XXXXX is presented', ['performance analysis', 'computational models'])\n", "('NONE', 'XXXXX in the atmega 128 of mica2 to compute the XXXXX', ['experimental analysis', 'execution time'])\n", "('method used for task', 'both XXXXX are useful in different contexts for XXXXX', ['hardware accelerators', 'risk analysis'])\n", "('method used for task', 'the XXXXX is found to provide better results than the XXXXX', ['svm classifier', 'pnn classifier'])\n", "('NONE', 'this study is the application of 2d linear XXXXX ( ca ) rules with the help of XXXXX to the problems of edge detection', ['cellular automata', 'fuzzy membership function'])\n", "('NONE', 'this study is the application of 2d linear XXXXX ( ca ) rules with the help of fuzzy membership function to the problems of XXXXX', ['cellular automata', 'edge detection'])\n", "('NONE', 'this study is the application of 2d linear cellular automata ( ca ) rules with the help of XXXXX to the problems of XXXXX', ['fuzzy membership function', 'edge detection'])\n", "('method used for task', 'an efficient and simple thresholding technique of XXXXX based on ca transition rules optimized by XXXXX ( pso ) is proposed', ['edge detection', 'particle swarm optimization method'])\n", "('NONE', 'coded XXXXX out performs uncoded system in terms of XXXXX', ['multi-carrier mimo system', 'ber performance'])\n", "('NONE', 'coded XXXXX with mmse/osic provides better XXXXX than system with coded zf/osic', ['multi-carrier mimo system', 'ber performance'])\n", "('NONE', 'a XXXXX using nsga-ii , entropy and XXXXX is employed', ['two-stage approach', 'topsis methods'])\n", "('NONE', 'a fully automatic system for XXXXX using single-spectral XXXXX is presented', ['tumor segmentation', 'mr images'])\n", "('NONE', 'a study for evaluating the efficacy of XXXXX over XXXXX is included', ['statistical features', 'gabor wavelet features'])\n", "('method used for task', 'randomized ftvbt elects tree nodes based on a XXXXX and increases XXXXX', ['weight function', 'network lifetime'])\n", "('method used for task', 'aecfv exhibits a high XXXXX , low false positive rate , faster attack detection , and lower XXXXX', ['detection rate', 'communication overhead'])\n", "('method used for task', 'data is intelligently analyzed using XXXXX ( hmm ) and XXXXX ( dtw )', ['hidden markov models', 'dynamic time warping'])\n", "('NONE', 'XXXXX determines the severity of the accident and reduces XXXXX', ['data analysis', 'false positives'])\n", "('NONE', 'a major XXXXX in synthesis of XXXXX is state assignment ( sa )', ['optimization problem', 'sequential circuits'])\n", "('NONE', 'a major XXXXX in synthesis of sequential circuits is XXXXX ( sa )', ['optimization problem', 'state assignment'])\n", "('method used for task', 'a major optimization problem in synthesis of XXXXX is XXXXX ( sa )', ['sequential circuits', 'state assignment'])\n", "('method used for task', 'the use of XXXXX microphone array for XXXXX and fusion is explored', ['data capture', 'kinect sensor'])\n", "('NONE', 'the XXXXX specifies the signal XXXXX', ['cost function', 'noise level'])\n", "('NONE', 'gamma equalization for XXXXX can improves the XXXXX', ['contrast enhancement', 'segmentation performance'])\n", "('NONE', 'to facilitate XXXXX , a XXXXX is proposed', ['confusion matrix', 'segmentation performance'])\n", "('method used for task', 'minimizes the XXXXX length , which , in turn , helps to reduce the XXXXX', ['test sequence', 'memory size'])\n", "('NONE', 'usage of a reduced XXXXX minimizes XXXXX during testing period', ['power dissipation', 'test sequence'])\n", "('NONE', 'by analysis of XXXXX of adjacent pixels as well as image interpolation techniques , a novel XXXXX is put forward', ['statistical properties', 'interpolation method'])\n", "('NONE', 'generally speaking , an assessment approach for the XXXXX situation by integrating the XXXXX with data fusion techniques is proposed', ['multi-agent system', 'rock burst'])\n", "('NONE', 'case study results shows the proposed XXXXX situation assessment model can give a relatively accurate forecast , and can help coal mine decision-makers to grasp an overview of the XXXXX of the XXXXX', ['rock burst', 'development trend'])\n", "('NONE', 'case study results shows the proposed XXXXX situation assessment model can give a relatively accurate forecast , and can help coal mine decision-makers to grasp an overview of the XXXXX of the XXXXX', ['development trend', 'rock burst'])\n", "('method used for task', 'the faster implementation of the XXXXX is useful during XXXXX of the 3d bio-medical images', ['affine transform', 'image registration'])\n", "('NONE', 'this work make a trade-off between XXXXX and its reliability associated with XXXXX and energy consumptions', ['modular design', 'fault tolerance'])\n", "('method used for task', 'refat enables run-time reconfiguration of XXXXX and XXXXX', ['network topology', 'routing algorithm'])\n", "('method used for task', 'XXXXX and XXXXX are reduced meanwhile area over-head is kept low', ['execution time', 'power consumption'])\n", "('NONE', 'an XXXXX for lifecycle energy consumption of XXXXX is proposed and evaluated', ['analytical model', 'mobile devices'])\n", "('NONE', 'a comparative XXXXX of performance and XXXXX of old and new android-based smartphone operating under the thin client paradigm was performed', ['empirical evaluation', 'energy consumption'])\n", "('NONE', 'a comparative XXXXX of performance and energy consumption of old and new android-based smartphone operating under the XXXXX was performed', ['empirical evaluation', 'thin client paradigm'])\n", "('NONE', 'a comparative empirical evaluation of performance and XXXXX of old and new android-based smartphone operating under the XXXXX was performed', ['energy consumption', 'thin client paradigm'])\n", "('method used for task', 'an XXXXX is then proposed to solve the XXXXX based on matrix distance', ['efficient algorithm', 'convex problem'])\n", "('NONE', 'framework tested in electrical XXXXX as a XXXXX', ['distribution systems', 'case study'])\n", "('NONE', 'we can do XXXXX with the XXXXX between acf ( t ) and a reference line', ['spectrum sensing', 'euclidean distance'])\n", "('method used for task', 'results show that XXXXX is preserved while XXXXX is enhanced', ['image brightness', 'image contrast'])\n", "('NONE', 'we predict XXXXX of embedded systems to extend their XXXXX', ['power consumption', 'service time'])\n", "('NONE', 'a XXXXX can directly achieve the XXXXX without scanning all channels', ['sensor node', 'l2 handover'])\n", "('method used for task', 'the XXXXX removes the ambiguity and improves the XXXXX', ['correlation function', 'tracking performance'])\n", "('method used for task', 'an XXXXX and a XXXXX of the system are presented', ['energy management', 'control strategy'])\n", "('NONE', 'we propose a game for reducing XXXXX in XXXXX', ['power consumption', 'data centers'])\n", "('method used for task', 'we establish XXXXX on the XXXXX of the game', ['upper bounds', 'convergence time'])\n", "('NONE', 'XXXXX , XXXXX & packet e2e delay are used to evaluate the protocols', ['packet loss', 'packet delivery'])\n", "('method used for task', 'applying weighted XXXXX to classify raw and decomposed XXXXX', ['permutation entropy', 'eeg signals'])\n", "('NONE', 'XXXXX denoising using nlm and XXXXX proposed', ['image sequence', 'zernike moments'])\n", "('method used for task', 'XXXXX is used to attain the more accurate XXXXX', ['sub-pixel accuracy', 'image alignment'])\n", "('NONE', 'XXXXX among different XXXXX', ['performance comparison', 'super-resolution algorithms'])\n", "('NONE', 'the XXXXX of the XXXXX ( cr ) network is improved , when malicious XXXXX user coexists in the network', ['cognitive radio', 'detection performance'])\n", "('NONE', 'the results indicate that for a constant detachment length , the XXXXX and surface effect are linked and it can be seen that the influence of XXXXX decreases with increasing XXXXX', ['beam thickness', 'surface effects'])\n", "('NONE', 'the results indicate that for a constant detachment length , the XXXXX and surface effect are linked and it can be seen that the influence of XXXXX decreases with increasing XXXXX', ['surface effects', 'beam thickness'])\n", "('NONE', 'XXXXX in terms of XXXXX is improved besides memory reduction obtained', ['image quality', 'signal-to-noise ratio'])\n", "('method used for task', 'decreasing XXXXX reduces system complexity and XXXXX', ['processing time', 'power consumption'])\n", "('NONE', 'we increase the XXXXX ratio ( pdr ) through the application of multi-channel technology in XXXXX', ['wireless sensor networks', 'packet delivery'])\n", "('method used for task', 'emd-bpnn has higher XXXXX and better XXXXX than standard bpnn and standard svr', ['prediction accuracy', 'generalization performance'])\n", "('method used for task', 'emd-bpnn can be used as a suitable and effective XXXXX for predicting XXXXX in intensive aquaculture', ['modeling tool', 'water temperature'])\n", "('method used for task', 'two pair-wise XXXXX are proposed for cdma based XXXXX', ['mac protocol', 'code assignment schemes'])\n", "('NONE', 'enhances the security of embedded message by using scene-change detection and discrete XXXXX of XXXXX', ['video sequences', 'cosine transforms'])\n", "('NONE', 'enhances XXXXX using discrete wavelet transforms in XXXXX', ['video quality', 'video sequences'])\n", "('NONE', 'social-spider optimization for XXXXX in XXXXX', ['model selection', 'support vector machines'])\n", "('method used for task', 'minimal added XXXXX needed to expand the score and avoid XXXXX', ['time complexity', 'information loss'])\n", "('method used for task', 'multi-level XXXXX is proposed to enhance XXXXX of lbp', ['discrimination power', 'quantization scheme'])\n", "('method used for task', 'a XXXXX is preprocessed to produce a XXXXX object and a blur image object', ['test image', 'noise image'])\n", "('method used for task', 'a XXXXX is preprocessed to produce a noise XXXXX and a blur XXXXX', ['test image', 'image object'])\n", "('method used for task', 'a test image is preprocessed to produce a XXXXX object and a blur XXXXX', ['noise image', 'image object'])\n", "('method used for task', 'a new energy efficient XXXXX has been proposed for cluster based XXXXX', ['routing algorithm', 'wireless sensor networks'])\n", "('method used for task', 'seed achieves o ( 1 ) message exchange complexity per XXXXX and XXXXX for cluster formation is o ( n ) for wsn having n sensor nodes', ['sensor node', 'time complexity'])\n", "('NONE', 'seed achieves o ( 1 ) XXXXX complexity per XXXXX and time complexity for cluster formation is o ( n ) for wsn having n sensor nodes', ['sensor node', 'message exchange'])\n", "('method used for task', 'seed achieves o ( 1 ) message exchange complexity per XXXXX and time complexity for cluster formation is o ( n ) for wsn having n XXXXX', ['sensor node', 'sensor nodes'])\n", "('method used for task', 'seed achieves o ( 1 ) XXXXX complexity per sensor node and XXXXX for cluster formation is o ( n ) for wsn having n sensor nodes', ['time complexity', 'message exchange'])\n", "('method used for task', 'seed achieves o ( 1 ) message exchange complexity per sensor node and XXXXX for cluster formation is o ( n ) for wsn having n XXXXX', ['time complexity', 'sensor nodes'])\n", "('method used for task', 'seed achieves o ( 1 ) XXXXX complexity per sensor node and time complexity for cluster formation is o ( n ) for wsn having n XXXXX', ['message exchange', 'sensor nodes'])\n", "('NONE', 'the computational accuracy of the propose method in the XXXXX and the overall XXXXX can reach up10−6–10−7', ['gaussian function', 'rbf nn'])\n", "('method used for task', 'the application of the vhdl code for a 3-5-1 XXXXX to the dynamic identification in the XXXXX and the pmsm drive system are successfully demonstrated', ['linear system', 'rbf nn'])\n", "('NONE', 'coded idma system with mutp based on msnr XXXXX regime results in a superior XXXXX with less snr when compared to the equal XXXXX policy', ['power allocation', 'ber performance'])\n", "('method used for task', 'we use a XXXXX to analyze the latency and XXXXX', ['probabilistic model', 'stability performances'])\n", "('NONE', 'the performance of hybrid XXXXX with median , median edge , gradient adjusted and local XXXXX is compared', ['local algorithm', 'prediction algorithms'])\n", "('NONE', 'the hybrid local XXXXX has the highest frequency of zero XXXXX', ['prediction algorithm', 'prediction error'])\n", "('NONE', 'the psnr and XXXXX are improved in the hybrid XXXXX', ['embedding capacity', 'local algorithm'])\n", "('method used for task', 'highlightanalyze the nonlinear cubic XXXXX ratio effect for ultrasonic XXXXX', ['stiffness coefficient', 'cutting system'])\n", "('method used for task', 'ultrasonic XXXXX is designed for stable and synchronized vibration when the excitation force is controlled at 2.0 with the interval of nonlinear cubic XXXXX ratio 0.006 ≦ η < 1.248', ['cutting system', 'stiffness coefficient'])\n", "('method used for task', 'filtering framework is based on XXXXX and XXXXX', ['temporal information', 'sparse representation'])\n", "('NONE', 'after a comprehensive comparison of sparse recovery algorithms , three were selected for our method : bayesian XXXXX ( bcs ) , bregman XXXXX , and orthogonal matching pursuit ( omp )', ['compressive sensing', 'iterative algorithm'])\n", "('method used for task', 'we survey the state scheduling-based XXXXX protocols suitable for unattended XXXXX', ['topology control', 'wireless sensor networks'])\n", "('method used for task', 'a new XXXXX extraction method is proposed based on XXXXX and distance regularized cv ( chan–vese ) model', ['image decomposition', 'river target'])\n", "('NONE', 'the mechanism relies on XXXXX , XXXXX and bit error rate', ['received signal strength', 'congestion level'])\n", "('NONE', 'evaluate the effectiveness of using cellular positioning techniques along with gps-sensor based XXXXX via performing experiments on XXXXX and location accuracy for various data acquisition modes where only gps positioning , only cellular positioning and both of the techniques are executed', ['data collection', 'energy consumption'])\n", "('method used for task', 'evaluate the effectiveness of using cellular positioning techniques along with gps-sensor based XXXXX via performing experiments on energy consumption and XXXXX for various data acquisition modes where only gps positioning , only cellular positioning and both of the techniques are executed', ['data collection', 'location accuracy'])\n", "('method used for task', 'evaluate the effectiveness of using cellular positioning techniques along with gps-sensor based data collection via performing experiments on XXXXX and XXXXX for various data acquisition modes where only gps positioning , only cellular positioning and both of the techniques are executed', ['energy consumption', 'location accuracy'])\n", "('NONE', 'our proposed system is designed to be used for average velocity calculation which is an essential parameter in traffic monitoring systems . therefore , we perform experiments to compute the average velocity by extracting XXXXX with the hybrid data acquisition model . we assume average speed measurements obtained from gps positioning as the XXXXX to assess the performance of our model', ['location data', 'ground truth data'])\n", "('NONE', 'assess the feasibility of cellular positioning in cases where the gps sensors are not available for location estimation . in this paper , we propose a hybrid acquisition model which applies cellular positioning techniques to obtain the raw XXXXX where gps sensor is not available or battery of a particular XXXXX is too low for location lookup', ['smart phone', 'location data'])\n", "('method used for task', 'when a XXXXX round is finished , an adaptive function is carried out to adjust the criteria evaluation in order to obtain a more adaptive ranking order of candidate XXXXX for the next round', ['data exchange', 'mobile devices'])\n", "('method used for task', 'when a XXXXX round is finished , an adaptive function is carried out to adjust the XXXXX in order to obtain a more adaptive ranking order of candidate mobile devices for the next round', ['data exchange', 'criteria evaluation'])\n", "('NONE', 'when a data exchange round is finished , an adaptive function is carried out to adjust the XXXXX in order to obtain a more adaptive ranking order of candidate XXXXX for the next round', ['mobile devices', 'criteria evaluation'])\n", "('method used for task', 'an XXXXX for routing XXXXX using recursive partition is proposed', ['efficient algorithm', 'multicast traffic'])\n", "('method used for task', 'a new XXXXX model to evaluate security risks associated with XXXXX on social pervasive applications', ['information sharing', 'risk indicator'])\n", "('NONE', 'a new risk indicator model to evaluate XXXXX associated with XXXXX on social pervasive applications', ['information sharing', 'security risks'])\n", "('method used for task', 'a new XXXXX model to evaluate XXXXX associated with information sharing on social pervasive applications', ['risk indicator', 'security risks'])\n", "('NONE', 'algorithm aims to improve the joint XXXXX in XXXXX based wban applications', ['signal recovery', 'compressed sensing'])\n", "('NONE', 'the XXXXX of sensing time and XXXXX of sus is achieved through novel iterative dinkelbach method ( nidm ) algorithm', ['joint optimization', 'transmission power'])\n", "('method used for task', 'the proposed XXXXX is compared with standard technologies and recent protocols in the literature , and it achieves better results for XXXXX , packet loss ratio , and throughput parameters', ['mac protocol', 'end-to-end delay'])\n", "('NONE', 'model the XXXXX by means of a XXXXX', ['context evolution', 'graph approach'])\n", "('method used for task', 'a XXXXX is proposed using the XXXXX', ['harmony search algorithm', 'feature selection approach'])\n", "('NONE', 'table-based routing can be used to implement XXXXX by reconfiguring table entries . the article shows how table entries can be computed efficiently , and how the XXXXX can be organized to function reliably even in presence of transmission errors', ['fault-tolerant routing', 'reconfiguration process'])\n", "('NONE', 'the additional hardware overhead for XXXXX table reconfiguration amounts to only 6 % of the XXXXX of a network switch', ['fault-tolerant routing', 'chip area'])\n", "('NONE', 'the coding XXXXX reaches 34 % , compared to the case of coding with no XXXXX', ['performance improvement', 'background suppression'])\n", "('NONE', 'detailed XXXXX of XXXXX , seig , and nonlinear load are presented', ['mathematical model', 'wind turbine'])\n", "('method used for task', 'research highlightsa multi-step XXXXX based on adaptive XXXXX generalized gradient vector flow using component-based normalization for snake model is proposed', ['decision model', 'edge preserving'])\n", "('method used for task', 'research highlightsa multi-step XXXXX based on adaptive edge preserving generalized gradient vector flow using component-based normalization for XXXXX is proposed', ['decision model', 'snake model'])\n", "('method used for task', 'research highlightsa multi-step decision model based on adaptive XXXXX generalized gradient vector flow using component-based normalization for XXXXX is proposed', ['edge preserving', 'snake model'])\n", "('NONE', 'the proposed algorithm presents a novel external force , which provides better results than other approaches in terms of XXXXX , weak XXXXX and convergence', ['noise robustness', 'edge preserving'])\n", "('method used for task', 'an improved multi-step XXXXX based on this novel external force is adopted , which adds new effective weighting function to attenuate the magnitudes of unwanted edges and adopts narrow band method to reduce XXXXX', ['decision model', 'time complexity'])\n", "('method used for task', 'the surface estimation problem is described as a max-margin based formulation of a XXXXX and solving the XXXXX using sub-gradient method', ['kernel function', 'objective function'])\n", "('NONE', 'the XXXXX problem is described as a max-margin based formulation of a XXXXX and solving the objective function using sub-gradient method', ['kernel function', 'surface estimation'])\n", "('method used for task', 'the XXXXX problem is described as a max-margin based formulation of a kernel function and solving the XXXXX using sub-gradient method', ['objective function', 'surface estimation'])\n", "('NONE', 'open issues and XXXXX of metaheuristics for XXXXX are addressed', ['future trends', 'healthcare system'])\n", "('NONE', 'a new modified grey level co-occurrence matrix ( mglcm ) method is presented to extract XXXXX statistical XXXXX for discriminating brain abnormality', ['second order', 'texture features'])\n", "('method used for task', 'mglcm generates efficient XXXXX that is used for measuring the symmetry of XXXXX scan than the tradition glcm', ['texture feature', 'mri brain'])\n", "('NONE', 'the XXXXX of XXXXX are optimized by genetic algorithm', ['membership functions', 'fuzzy controller'])\n", "('NONE', 'the XXXXX of fuzzy controller are optimized by XXXXX', ['membership functions', 'genetic algorithm'])\n", "('NONE', 'the membership functions of XXXXX are optimized by XXXXX', ['fuzzy controller', 'genetic algorithm'])\n", "('NONE', 'the proposed algorithm improves XXXXX even with long XXXXX', ['convergence speed', 'adaptive filters'])\n", "('NONE', 'the led lighting strip is controlled by the XXXXX via a wi-fi XXXXX', ['mobile app', 'wireless network'])\n", "('NONE', 'the XXXXX was performed by using XXXXX ( svm ) and k nearest neighbors ( knn )', ['emotion classification', 'support vector machine'])\n", "('method used for task', 'sensitivity and XXXXX for gis-multicriteria XXXXX is underdeveloped in giscience', ['uncertainty analysis', 'decision making'])\n", "('method used for task', 'XXXXX included for XXXXX from matching keypoints', ['outlier removal', 'ransac algorithm'])\n", "('method used for task', 'we propose a simple and effective XXXXX for XXXXX', ['3d meshes', 'partitioning algorithm'])\n", "('NONE', 'XXXXX are estimated accurately using new XXXXX', ['camera parameters', 'energy function'])\n", "('NONE', 'consider the influence of the bond condition on the XXXXX of XXXXX', ['reinforced concrete beams', 'fire resistance'])\n", "('NONE', 'assess the integrity of XXXXX under XXXXX', ['reinforced concrete beams', 'fire conditions'])\n", "('method used for task', 'XXXXX and XXXXX of local beam sections', ['numerical analysis', 'parametric study'])\n", "('NONE', 'simulation of complex XXXXX with non-homogeneous XXXXX', ['flow problems', 'material parameters'])\n", "('NONE', 'a XXXXX for analysis of virtual XXXXX is developed', ['conceptual framework', 'supply chains'])\n", "('method used for task', 'development of the agribusiness XXXXX based on analytic XXXXX', ['location model', 'hierarchy process'])\n", "('NONE', 'XXXXX with exponential weight-criterion XXXXX and constraints', ['genetic algorithm', 'fitness function'])\n", "('method used for task', 'the XXXXX is XXXXX and easily wearable', ['hybrid system', 'low cost'])\n", "('method used for task', 'results of XXXXX are compared to results of various XXXXX', ['oxygen saturation', 'clinical trials'])\n", "('NONE', 'we design patient-specific XXXXX shapes using XXXXX', ['topology optimization', 'bone replacement'])\n", "('NONE', 'the algorithm denoise XXXXX recorded in noisy environments by XXXXX', ['mobile devices', 'pcg signals'])\n", "('NONE', 'planar 2d XXXXX are in good agreement with complex XXXXX', ['numerical models', '3d models'])\n", "('method used for task', 'presenting a new two-stage meta-heuristic XXXXX based on general XXXXX', ['clustering algorithm', 'type-2 fuzzy sets'])\n", "('NONE', 'incorporating a new similarity-based XXXXX using alpha-plane representation of general XXXXX', ['objective function', 'type-2 fuzzy sets'])\n", "('NONE', 'XXXXX about 8-ms for detection of the XXXXX', ['response time', 'qrs complexes'])\n", "('method used for task', 'classifiers methodologies used included XXXXX and lasso XXXXX', ['genetic algorithms', 'logistic regression'])\n", "('NONE', 'entropies , hos , fd and XXXXX are extracted from XXXXX', ['fundus images', 'gabor wavelet features'])\n", "('method used for task', 'XXXXX provide an opportunity to improve XXXXX', ['electronic medical records', 'patient care'])\n", "('NONE', 'XXXXX : XXXXX and fuzzy connectedness is employed', ['soft computing', 'evolutionary computation'])\n", "('method used for task', 'XXXXX : evolutionary computation and XXXXX is employed', ['soft computing', 'fuzzy connectedness'])\n", "('method used for task', 'soft computing : XXXXX and XXXXX is employed', ['evolutionary computation', 'fuzzy connectedness'])\n", "('NONE', 'model discovery by large scale XXXXX coupled with XXXXX', ['feature extraction', 'machine learning'])\n", "('method used for task', 'a XXXXX for glucose utilization is proposed based on XXXXX', ['global model', 'time-varying parameters'])\n", "('method used for task', 'patient-specific XXXXX for XXXXX are proposed to avoid leakage', ['graph cuts', 'shape constraints'])\n", "('NONE', 'analysis of XXXXX in 62 consecutive patients with XXXXX ( af )', ['atrial fibrillation', 'surface ecg'])\n", "('NONE', 'in this paper , bel is applied for the XXXXX of gene-expression XXXXX', ['microarray data', 'classification tasks'])\n", "('NONE', 'proposed method improves the XXXXX of srbct , hgg and XXXXX', ['detection accuracy', 'lung cancer'])\n", "('method used for task', 'we calculate the XXXXX between the adjusted endpoints to find the accurate XXXXX', ['shortest path', 'splitting line'])\n", "('method used for task', 'we present an analysis of XXXXX and XXXXX applied to odor discrimination', ['pattern recognition', 'dimensionality reduction techniques'])\n", "('method used for task', 'proposing a XXXXX for pediatric XXXXX', ['decision support system', 'epilepsy diagnosis'])\n", "('NONE', 'evaluation of graph theory measures of XXXXX in pediatric XXXXX', ['brain functional connectivity', 'epilepsy diagnosis'])\n", "('method used for task', 'XXXXX to investigate XXXXX and tracking convergence', ['lyapunov stability theorem', 'global stability'])\n", "('method used for task', 'many gene selection methods with XXXXX and XXXXX increase the accuracy of autism recognition', ['genetic algorithm', 'svm classifiers'])\n", "('NONE', 'XXXXX can measure complexity of clinical XXXXX', ['permutation entropy', 'time series'])\n", "('method used for task', 'we compared XXXXX to XXXXX', ['logistic regression models', 'data mining algorithms'])\n", "('NONE', 'the XXXXX performed as well as a simple XXXXX', ['logistic regression model', 'data mining algorithms'])\n", "('NONE', 'application 2 : investigation of XXXXX ( XXXXX of the action potentials along the fibres )', ['muscle activity', 'conduction velocity'])\n", "('NONE', 'application 2 : investigation of XXXXX ( conduction velocity of the XXXXX along the fibres )', ['muscle activity', 'action potentials'])\n", "('NONE', 'application 2 : investigation of muscle activity ( XXXXX of the XXXXX along the fibres )', ['conduction velocity', 'action potentials'])\n", "('NONE', 'a XXXXX of the XXXXX is presented', ['parametric model', 'left ventricle'])\n", "('method used for task', 'this is the first study to validate skin XXXXX to a XXXXX', ['gold standard', 'thickness estimation'])\n", "('NONE', 'XXXXX resolves XXXXX in simulation of neural systems', ['discontinuous galerkin fem', 'stability problem'])\n", "('NONE', 'XXXXX excels XXXXX for lif and lifb models', ['monte carlo simulation', 'discontinuous galerkin fem'])\n", "('NONE', 'a system-level XXXXX of chronic high XXXXX effects is conducted', ['computer simulation', 'sodium intake'])\n", "('method used for task', 'modeling results suggest high XXXXX alters the body-fluid homeostasis and increases the lv XXXXX and relative wall thickness', ['sodium intake', 'work load'])\n", "('method used for task', 'modeling results suggest high XXXXX alters the body-fluid homeostasis and increases the lv work load and relative XXXXX', ['sodium intake', 'wall thickness'])\n", "('method used for task', 'modeling results suggest high sodium intake alters the body-fluid homeostasis and increases the lv XXXXX and relative XXXXX', ['work load', 'wall thickness'])\n", "('method used for task', 'first implementation of a XXXXX with dendritic processing to solve classify XXXXX', ['retinal images', 'lattice neural network'])\n", "('NONE', 'end-tidal XXXXX decreased when a fall in XXXXX diagnosed', ['carbon dioxide', 'cardiac output'])\n", "('NONE', 'multiple expert readers produced a XXXXX of 430 abstracts on XXXXX', ['gold standard', 'lung cancer'])\n", "('NONE', 'we determine the value of XXXXX using 3 disparate XXXXX', ['free text', 'use cases'])\n", "('method used for task', 'XXXXX specific values and limitations are identified in XXXXX', ['use case', 'large data sets'])\n", "('NONE', 'XXXXX are represented by a statistical function with XXXXX', ['normal distribution', 'tissue properties'])\n", "('NONE', 'mean value of the XXXXX are identified using XXXXX', ['genetic algorithm', 'material parameters'])\n", "('NONE', 'using forward XXXXX with XXXXX', ['feature selection', 'svm classification'])\n", "('NONE', 'the state-of-the-art research in XXXXX of XXXXX is presented', ['automated diagnosis', 'celiac disease'])\n", "('NONE', 'we analyze the XXXXX through the XXXXX', ['load distribution', 'wrist joint'])\n", "('method used for task', 'we used the XXXXX model method on a XXXXX of the wrist joint', ['3d model', 'rigid body spring'])\n", "('NONE', 'we used the rigid body spring model method on a XXXXX of the XXXXX', ['3d model', 'wrist joint'])\n", "('NONE', 'we used the XXXXX model method on a 3d model of the XXXXX', ['rigid body spring', 'wrist joint'])\n", "('NONE', 'XXXXX distinguish the three types of XXXXX', ['statistical features', 'eye movements'])\n", "('NONE', 'the proposed XXXXX show superior XXXXX compared to traditional features', ['geometric features', 'classification performance'])\n", "('NONE', 'a novel method for XXXXX of XXXXX ( af ) is proposed', ['automatic detection', 'atrial fibrillation'])\n", "('NONE', 'linking intrinsic features of XXXXX with the optimal XXXXX that maximizes the similarity entropy', ['time series', 'pattern size'])\n", "('NONE', 'the method is built upon XXXXX of XXXXX', ['information fusion', 'endpoint responses'])\n", "('NONE', 'XXXXX trained and tested on a large sample of real XXXXX', ['prediction algorithm', 'clinical data'])\n", "('NONE', 'we applied a XXXXX algorithm-based XXXXX', ['machine learning', 'classification method'])\n", "('method used for task', 'introduced model could be applied in XXXXX and XXXXX', ['ischemia monitoring', 'tumor detection'])\n", "('NONE', 'the proposed system achieves high XXXXX in an XXXXX', ['detection accuracy', 'empirical study'])\n", "('NONE', 'the XXXXX incorporates the edges obtained using a tunable XXXXX', ['cost function', 'gabor filter'])\n", "('NONE', 'the XXXXX depends on XXXXX of matrix fibrous component', ['topological properties', 'healing process'])\n", "('method used for task', 'we propose to accelerate pls–rfe based XXXXX to classify XXXXX', ['gene selection', 'microarray data'])\n", "('method used for task', 'we developed an XXXXX to automatically analyze XXXXX', ['image processing pipeline', 'muscle fibers'])\n", "('NONE', '3d-rsd is fast and more accurate than three other reference methods , and can be used in high-content screening tasks to evaluate XXXXX in human XXXXX lines', ['drug efficacy', 'cancer cell'])\n", "('NONE', 'automatic XXXXX from XXXXX', ['feature extraction', 'retinal images'])\n", "('method used for task', 'XXXXX is common after XXXXX and it is a serious health problem worldwide', ['ventricular tachycardia', 'myocardial infarction'])\n", "('method used for task', 'XXXXX is common after myocardial infarction and it is a serious XXXXX worldwide', ['ventricular tachycardia', 'health problem'])\n", "('method used for task', 'ventricular tachycardia is common after XXXXX and it is a serious XXXXX worldwide', ['myocardial infarction', 'health problem'])\n", "('method used for task', 'the source of reentrant XXXXX is often located in the infarct XXXXX , the layer of thin surviving myocardium adjacent to the infarct', ['ventricular tachycardia', 'border zone'])\n", "('NONE', 'although there is electrical activity occurring in the infarct XXXXX , the XXXXX of the leading edge of the propagating wavefront is not constant', ['conduction velocity', 'border zone'])\n", "('method used for task', 'in this study , the activation rate and ibz XXXXX necessary for functional block to form during premature stimulation and during reentrant XXXXX , are determined', ['geometric structure', 'ventricular tachycardia'])\n", "('method used for task', 'in this study , the XXXXX and ibz XXXXX necessary for functional block to form during premature stimulation and during reentrant ventricular tachycardia , are determined', ['geometric structure', 'activation rate'])\n", "('method used for task', 'in this study , the XXXXX and ibz geometric structure necessary for functional block to form during premature stimulation and during reentrant XXXXX , are determined', ['ventricular tachycardia', 'activation rate'])\n", "('NONE', 'the XXXXX manipulates the cardiac XXXXX to suppress alternans', ['control algorithm', 'tissue mechanics'])\n", "('method used for task', 'improving the reasoning in XXXXX for better XXXXX', ['conceptual graphs', 'knowledge representation'])\n", "('NONE', 'XXXXX of age-related macular degeneration ( amd ) using XXXXX', ['automated detection', 'fundus images'])\n", "('NONE', 'an XXXXX angiogenic process is modeled by a suitable version of the cellular XXXXX', ['in vivo', 'potts model'])\n", "('NONE', 'by detecting associations with XXXXX , XXXXX can be discovered', ['celiac disease', 'research trends'])\n", "('NONE', 'XXXXX were mainly enriched in XXXXX and ribosome pathway', ['feature genes', 'cell proliferation'])\n", "('NONE', 'XXXXX with multimodal data can accurately predict postsurgical outcome in patients with drug resistant mesial XXXXX', ['machine learning', 'temporal lobe epilepsy'])\n", "('method used for task', 'a XXXXX based on the quantitative XXXXX was developed to classify breast masses', ['computer-aided diagnosis system', 'strain features'])\n", "('method used for task', 'we used the combination of XXXXX ( svm ) and improved XXXXX ( iaco )', ['support vector machine', 'ant colony optimization'])\n", "('NONE', 'sbdne is applied to XXXXX using XXXXX', ['cancer classification', 'gene expression data'])\n", "('NONE', 'exploration of 14 XXXXX ( 86 XXXXX )', ['color spaces', 'color features'])\n", "('NONE', 'XXXXX of XXXXX for disease gene and biomarker discovery', ['automated analysis', 'high throughput data'])\n", "('method used for task', 'a novel computer-aided XXXXX for XXXXX using retinal fundus images is proposed', ['decision support system', 'diabetic retinopathy'])\n", "('NONE', 'a novel computer-aided XXXXX for diabetic retinopathy using retinal XXXXX is proposed', ['decision support system', 'fundus images'])\n", "('NONE', 'a novel computer-aided decision support system for XXXXX using retinal XXXXX is proposed', ['diabetic retinopathy', 'fundus images'])\n", "('method used for task', 'ten XXXXX are compared using stability and XXXXX', ['similarity measures', 'feature selection methods'])\n", "('method used for task', 'fs yielding the highest XXXXX are mrmr and XXXXX', ['bhattacharyya distance', 'prediction performance'])\n", "('method used for task', 'the proposed XXXXX is used to control a real three joints robot arm to grasp a XXXXX', ['bci system', 'target object'])\n", "('NONE', 'the controllability of our bci system is improved with the XXXXX embedded in the XXXXX', ['feedback mechanism', 'robot arm'])\n", "('method used for task', 'the controllability of our XXXXX is improved with the XXXXX embedded in the robot arm', ['feedback mechanism', 'bci system'])\n", "('method used for task', 'the controllability of our XXXXX is improved with the feedback mechanism embedded in the XXXXX', ['robot arm', 'bci system'])\n", "('method used for task', 'XXXXX is common after XXXXX', ['ventricular tachycardia', 'myocardial infarction'])\n", "('method used for task', 'reentrant XXXXX is often located in the infarct XXXXX', ['ventricular tachycardia', 'border zone'])\n", "('NONE', 'activation rate and ibz XXXXX during reentrant XXXXX are determined', ['geometric structure', 'ventricular tachycardia'])\n", "('method used for task', 'XXXXX and ibz XXXXX during reentrant ventricular tachycardia are determined', ['geometric structure', 'activation rate'])\n", "('method used for task', 'XXXXX and ibz geometric structure during reentrant XXXXX are determined', ['ventricular tachycardia', 'activation rate'])\n", "('NONE', 'we show how to store intrinsic features from XXXXX in a XXXXX', ['medical images', 'data warehouse'])\n", "('NONE', 'an improved XXXXX based thalamic XXXXX was proposed', ['fuzzy connectedness', 'segmentation method'])\n", "('method used for task', 'investigating a broad range of XXXXX for XXXXX', ['machine learning techniques', 'statistical analyses'])\n", "('NONE', 'evaluation of XXXXX from geometry independent of XXXXX', ['material properties', 'bone strength'])\n", "('method used for task', 'its components are implemented as XXXXX and supported by XXXXX', ['web services', 'cloud services'])\n", "('NONE', 'XXXXX as XXXXX to leverage statistics of the lesion', ['mutual information', 'energy function'])\n", "('NONE', 'three major registration frameworks were examined : ( a ) intensity-based , exhaustive XXXXX using three distinct XXXXX ( b ) geometry-based XXXXX , featuring three geometrical descriptors ( c ) the original implementation of the iterative closest point algorithm', ['cost functions', 'registration framework'])\n", "('NONE', 'three major registration frameworks were examined : ( a ) intensity-based , exhaustive XXXXX using three distinct XXXXX ( b ) geometry-based XXXXX , featuring three geometrical descriptors ( c ) the original implementation of the iterative closest point algorithm', ['cost functions', 'registration framework'])\n", "('method used for task', 'use of geometrical XXXXX for aligning 3d XXXXX', ['feature descriptors', 'medical data'])\n", "('method used for task', 'the devised XXXXX is validated with XXXXX', ['finite element analysis', 'force system'])\n", "('NONE', 'the XXXXX uses a bio-inspired XXXXX on the hermite transform to code local image features', ['optical flow estimation', 'model based'])\n", "('method used for task', 'the XXXXX uses a bio-inspired model based on the hermite transform to code local XXXXX', ['optical flow estimation', 'image features'])\n", "('method used for task', 'the optical flow estimation uses a bio-inspired XXXXX on the hermite transform to code local XXXXX', ['model based', 'image features'])\n", "('NONE', 'the XXXXX allows an unprecedented XXXXX in cell dynamics', ['wavelet transform', 'multi-scale analysis'])\n", "('method used for task', 'we used XXXXX and XXXXX for wheeze recognition', ['short-time fourier transform', 'svm classifier'])\n", "('NONE', 'saccadic XXXXX could be detected efficiently by an XXXXX', ['eye movements', 'adaptive algorithm'])\n", "('method used for task', 'prediction with XXXXX is comparable to multivariate XXXXX', ['random forest', 'linear regression'])\n", "('method used for task', 'the textural XXXXX is extracted by using discrete XXXXX and fractal technology', ['feature set', 'curvelet transform'])\n", "('method used for task', 'the ga–svm method is proposed for XXXXX and wce XXXXX', ['feature selection', 'image classification'])\n", "('NONE', 'XXXXX also reduces the peak von mises stress in the XXXXX and porous cage', ['cortical bone', 'bone fusion'])\n", "('method used for task', 'we proposed a novel gap-search XXXXX ( mrf ) for accurate cervical smear XXXXX', ['markov random field', 'image segmentation'])\n", "('method used for task', 'the method outperforms XXXXX and XXXXX', ['uniform sampling', 'compressive sensing'])\n", "('NONE', 'XXXXX with XXXXX and adaptive genetic operators is proposed', ['hybrid algorithm', 'cuckoo search'])\n", "('NONE', 'XXXXX with cuckoo search and adaptive XXXXX is proposed', ['hybrid algorithm', 'genetic operators'])\n", "('method used for task', 'hybrid algorithm with XXXXX and adaptive XXXXX is proposed', ['cuckoo search', 'genetic operators'])\n", "('method used for task', 'XXXXX showed that omc deposition was highly sensitive to particle charge and size and less sensitive to the inhalation XXXXX', ['sensitivity analysis', 'flow rate'])\n", "('NONE', 'XXXXX of whole body spine is segmented using XXXXX', ['ct images', 'active contour method'])\n", "('method used for task', 'fully automated XXXXX for XXXXX is proposed', ['active contour method', 'initialization method'])\n", "('NONE', 'voxel-classification-based segmentation suffers from XXXXX of XXXXX', ['high dimensionality', 'mr images'])\n", "('method used for task', 'a XXXXX for automatic XXXXX in colonoscopy videos is introduced', ['model based method', 'inflammation detection'])\n", "('method used for task', 'the proposed method is suitable for XXXXX and XXXXX of high-resolution colonoscopy videos', ['parallel implementation', 'real-time processing'])\n", "('NONE', 'XXXXX and state-of-the XXXXX allowed accurate segmentation', ['texture descriptors', 'art features'])\n", "('method used for task', 'a cs framework of XXXXX is proposed for XXXXX ( mecg ) signals in eigenspace', ['data reduction', 'multichannel ecg'])\n", "('NONE', 'pca is used to exploit the XXXXX across the channels resulting into sparse XXXXX', ['spatial correlation', 'eigenspace signals'])\n", "('method used for task', 'using the XXXXX ( cs ) approach , the significant eigenspace signals are gone through further XXXXX', ['compressed sensing', 'dimensionality reduction'])\n", "('NONE', 'using the XXXXX ( cs ) approach , the significant XXXXX are gone through further dimensionality reduction', ['compressed sensing', 'eigenspace signals'])\n", "('NONE', 'using the compressed sensing ( cs ) approach , the significant XXXXX are gone through further XXXXX', ['dimensionality reduction', 'eigenspace signals'])\n", "('NONE', 'a feature selection approach for the large number of XXXXX calculated by XXXXX', ['sparse representation', 'dictionary learning'])\n", "('NONE', 'a XXXXX for the large number of XXXXX calculated by dictionary learning', ['sparse representation', 'feature selection approach'])\n", "('NONE', 'a XXXXX for the large number of sparse representation calculated by XXXXX', ['dictionary learning', 'feature selection approach'])\n", "('NONE', 'XXXXX was carried out by using XXXXX', ['automatic segmentation', 'image processing methods'])\n", "('method used for task', 'statistics of gray levels , XXXXX , and XXXXX were computed', ['texture features', 'shape features'])\n", "('NONE', 'XXXXX of normal and amd classes using XXXXX', ['automated detection', 'fundus images'])\n", "('method used for task', 'XXXXX and XXXXX are used for feature extraction', ['radon transform', 'discrete wavelet transform'])\n", "('method used for task', 'XXXXX and discrete wavelet transform are used for XXXXX', ['radon transform', 'feature extraction'])\n", "('method used for task', 'radon transform and XXXXX are used for XXXXX', ['discrete wavelet transform', 'feature extraction'])\n", "('NONE', 'XXXXX with higher XXXXX affect p–j and p–o fits', ['self-presentation messages', 'argument quality'])\n", "('method used for task', 'interfacing with a XXXXX ( vs. a text screen ) augments efforts to establish XXXXX', ['human body', 'common ground'])\n", "('method used for task', 'XXXXX were more pronounced on the XXXXX', ['practice effects', 'memory measures'])\n", "('method used for task', 'fraping is an accepted social norm amongst some XXXXX , yet is frowned upon by XXXXX', ['young adults', 'older adults'])\n", "('method used for task', 'a XXXXX shows that a sealed bid environment is beneficial to a XXXXX', ['computer simulation', 'search engine'])\n", "('method used for task', 'a XXXXX is studied to reduce the XXXXX of candidate ensembles', ['search space', 'selection strategy'])\n", "('NONE', 'XXXXX increases the interest of students in XXXXX', ['virtual environment', '3d modelling'])\n", "('method used for task', 'overall XXXXX ( osr ) was chosen as the metrics for XXXXX', ['performance evaluation', 'success rate'])\n", "('method used for task', 'analysis and novel architecture for new XXXXX for wireless XXXXX', ['industry standard', 'power transfer'])\n", "('method used for task', 'a fast XXXXX ( mst ) -based clustering methodology that is robust to density-based outliers is proposed for large XXXXX', ['minimum spanning tree', 'high-dimensional data sets'])\n", "('NONE', 'the performance of the proposed algorithms is demonstrated on a number of small low-dimensional and large XXXXX with respect to several state-of-the-art XXXXX', ['clustering algorithms', 'high-dimensional data sets'])\n", "('NONE', 'problem of causal relations in XXXXX among XXXXX is studied', ['frequency domain', 'time series'])\n", "('NONE', 'problem of XXXXX in XXXXX among time series is studied', ['frequency domain', 'causal relations'])\n", "('NONE', 'problem of XXXXX in frequency domain among XXXXX is studied', ['time series', 'causal relations'])\n", "('NONE', 'the XXXXX of the proposed algorithm is compared with dwt and XXXXX', ['resource utilization', 'kalman filter'])\n", "('NONE', 'the ffm-aukf method provides accurate results in the XXXXX of non-linear XXXXX in presence of maneuver', ['state estimation', 'dynamic systems'])\n", "('method used for task', 'XXXXX for at XXXXX', ['channel equalization', 'fading channels'])\n", "('method used for task', 'awi provides good XXXXX and requires a low XXXXX', ['computational cost', 'reconstruction performance'])\n", "('method used for task', 'our algorithm ensures XXXXX and XXXXX of vector map', ['access control', 'copyright protection'])\n", "('NONE', 'our algorithm ensures XXXXX and copyright protection of XXXXX', ['access control', 'vector map'])\n", "('NONE', 'our algorithm ensures access control and XXXXX of XXXXX', ['copyright protection', 'vector map'])\n", "('method used for task', 'an overview is provided of key XXXXX for group and extended XXXXX', ['sequential monte carlo methods', 'object tracking'])\n", "('NONE', 'dna watermarking is for XXXXX and authentication of a XXXXX', ['copyright protection', 'dna sequence'])\n", "('NONE', 'there are n degrees of freedom for designing the XXXXX by using the XXXXX', ['discrete fourier transform', 'mask coefficients'])\n", "('NONE', 'design of the mask coefficients are formulated as an XXXXX with an XXXXX nonconvex function', ['optimization problem', 'l1 norm'])\n", "('NONE', 'design of the XXXXX are formulated as an XXXXX with an l1 norm nonconvex function', ['optimization problem', 'mask coefficients'])\n", "('NONE', 'design of the XXXXX are formulated as an optimization problem with an XXXXX nonconvex function', ['l1 norm', 'mask coefficients'])\n", "('NONE', 'a characterization of the XXXXX corresponding to weighted orthogonal constrained ica algorithms using symmetric XXXXX weighted unitary mapping is obtained', ['stationary points', 'minimum distance'])\n", "('NONE', 'the monotonic XXXXX of the weighted orthogonal constrained fixed point ica algorithms using symmetric XXXXX weighted unitary mapping for convex contrast function is proved , which is further extended to nonconvex contrast function case', ['convergence property', 'minimum distance'])\n", "('method used for task', 'we present a XXXXX for graph analysis that can be used for automatic XXXXX and descriptor evaluation', ['ranking function', 'feature selection'])\n", "('method used for task', 'the goal is to build an XXXXX where descriptors can be analysed for automatic XXXXX', ['evaluation framework', 'feature selection'])\n", "('NONE', '3-dimensional XXXXX in ocean with XXXXX', ['source localization', 'non-gaussian noise'])\n", "('NONE', 'automatic parameter determination in XXXXX based XXXXX', ['sparse representation', 'face hallucination'])\n", "('method used for task', 'XXXXX is analytically tractable under XXXXX', ['regularization parameter', 'map framework'])\n", "('method used for task', 'this new approach has given two new algorithms for XXXXX and XXXXX', ['noise reduction', 'speech enhancement'])\n", "('NONE', 'feedback plant identification is formulated as a XXXXX based XXXXX', ['cuckoo search', 'optimization problem'])\n", "('method used for task', 'XXXXX shows enhanced identification accuracy for XXXXX', ['simulation study', 'cuckoo search'])\n", "('NONE', 'the proposed XXXXX provides excellent XXXXX on noisy images', ['similarity measure', 'recognition performance'])\n", "('method used for task', 'a distributed XXXXX is designed with time delays and XXXXX', ['fusion filter', 'packet losses'])\n", "('NONE', 'a XXXXX for existence of steady-state XXXXX is given', ['sufficient condition', 'fusion filter'])\n", "('method used for task', 'we employ an XXXXX to initialize the XXXXX', ['expectation-maximization algorithm', 'model parameters'])\n", "('NONE', 'according to random finite set ( rfs ) and information inequality , the paper derives the joint detection and estimation ( jde ) XXXXX of an unresolved target-group using single and XXXXX in the presence of missed detection and clutter', ['error bounds', 'multiple sensors'])\n", "('NONE', 'the XXXXX provide important indications of the XXXXX for existing unresolved target-group jde approaches', ['error bounds', 'performance limitations'])\n", "('method used for task', 'inherent mechanisms of the cell-like XXXXX are utilized to realize an efficient and robust XXXXX', ['p system', 'multi-level thresholding method'])\n", "('NONE', 'evaluation of the XXXXX with analysis , XXXXX and experiment', ['numerical simulation', 'delay network'])\n", "('method used for task', 'we present a fast and XXXXX for XXXXX', ['efficient algorithm', 'image inpainting'])\n", "('NONE', 'we present a method for estimating a time-scale local XXXXX on XXXXX', ['hurst exponent', 'time series'])\n", "('method used for task', 'XXXXX is addressed using XXXXX', ['camera model identification', 'hypothesis testing theory'])\n", "('method used for task', 'a novel XXXXX based on the colour and spatial histograms over the XXXXX', ['similarity measure', 'lattice structure'])\n", "('method used for task', 'there is no study describing 2d XXXXX based on XXXXX in the literature', ['metaheuristic algorithms', 'adaptive filter algorithm'])\n", "('method used for task', 'we propose two novel XXXXX for unsupervised non-gaussian XXXXX', ['feature selection', 'inference frameworks'])\n", "('method used for task', 'determination of the optimal XXXXX to minimize the false-alarm without decreasing XXXXX', ['additive noise', 'detection probability'])\n", "('method used for task', 'XXXXX based on XXXXX and frequency warping operator', ['compressive sensing', 'wavelet analysis'])\n", "('method used for task', 'introduce the XXXXX which is more efficient than the XXXXX to fcm based algorithm', ['mahalanobis distance', 'euclidean distance'])\n", "('method used for task', 'combine the XXXXX and the XXXXX with the fcm algorithm based on kl information', ['mahalanobis distance', 'regularization term'])\n", "('NONE', 'combine the XXXXX and the regularization term with the XXXXX based on kl information', ['mahalanobis distance', 'fcm algorithm'])\n", "('NONE', 'combine the mahalanobis distance and the XXXXX with the XXXXX based on kl information', ['regularization term', 'fcm algorithm'])\n", "('method used for task', 'the XXXXX for fda XXXXX is derived with frequency increment error', ['data model', 'mimo radar'])\n", "('NONE', 'the subspace-based XXXXX of the fda XXXXX is analyzed', ['mimo radar', 'localization performance'])\n", "('NONE', 'XXXXX describing the algorithm behavior in XXXXX are obtained ;', ['steady state', 'model expressions'])\n", "('NONE', 'the new XXXXX with the p-norm XXXXX of the error signal is presented', ['cost function', 'logarithmic transformation'])\n", "('method used for task', 'dihs algorithm for improving the XXXXX and XXXXX', ['solution quality', 'search efficiency'])\n", "('method used for task', 'take-one strategy for a XXXXX to globally XXXXX is analyzed', ['fast convergence', 'optimal solution'])\n", "('method used for task', 'a new XXXXX based on new adaptive XXXXX is proposed', ['filtering process', 'diffusion function'])\n", "('method used for task', 'estimate the XXXXX based on signal inpainting and XXXXX', ['sparse representation', 'voltage fluctuation'])\n", "('NONE', 'develop the XXXXX on the recover error of the XXXXX component', ['upper bound', 'voltage fluctuation'])\n", "('method used for task', 'XXXXX is addressed using XXXXX', ['camera model identification', 'hypothesis testing theory'])\n", "('NONE', 'the msd-optimal XXXXX achieves the fastest XXXXX on every iteration', ['step size', 'convergence rate'])\n", "('method used for task', 'a new XXXXX rbp is presented for XXXXX in this paper', ['image descriptor', 'face recognition'])\n", "('method used for task', 'method based on periodic–chaotic states of the XXXXX for XXXXX', ['duffing oscillator', 'time-frequency analysis'])\n", "('NONE', 'the magnetic XXXXX can control the stroke of the pin accurately without any XXXXX', ['feedback control', 'latch mechanism'])\n", "('NONE', 'a high XXXXX can be achieved by the proposed XXXXX', ['contrast ratio', 'pixel circuit'])\n", "('NONE', 'application of organic transparent electrodes based on XXXXX in XXXXX', ['conducting polymers', 'display devices'])\n", "('NONE', 'ramp XXXXX by XXXXX and temperature is analyzed', ['slope variation', 'image load'])\n", "('NONE', 'XXXXX caused by XXXXX is compensated for by redesigning feedback current path in the driver circuit', ['slope variation', 'image load'])\n", "('method used for task', 'XXXXX caused by image load is compensated for by redesigning feedback current path in the XXXXX', ['slope variation', 'driver circuit'])\n", "('method used for task', 'slope variation caused by XXXXX is compensated for by redesigning feedback current path in the XXXXX', ['image load', 'driver circuit'])\n", "('NONE', 'XXXXX was induced by intermolecular XXXXX from eu ( iii ) ion to viologen', ['energy transfer', 'emission control'])\n", "('NONE', 'a novel 7t1c pixel circuit is proposed for XXXXX , high XXXXX , and low power amoled displays', ['high resolution', 'frame rate'])\n", "('method used for task', 'a novel 7t1c pixel circuit is proposed for XXXXX , high frame rate , and XXXXX amoled displays', ['high resolution', 'low power'])\n", "('method used for task', 'a novel 7t1c XXXXX is proposed for XXXXX , high frame rate , and low power amoled displays', ['high resolution', 'pixel circuit'])\n", "('method used for task', 'a novel 7t1c pixel circuit is proposed for high resolution , high XXXXX , and XXXXX amoled displays', ['frame rate', 'low power'])\n", "('method used for task', 'a novel 7t1c XXXXX is proposed for high resolution , high XXXXX , and low power amoled displays', ['frame rate', 'pixel circuit'])\n", "('method used for task', 'a novel 7t1c XXXXX is proposed for high resolution , high frame rate , and XXXXX amoled displays', ['low power', 'pixel circuit'])\n", "('NONE', 'the results show that a small depth ( XXXXX of 2.3–3.9arc min ) allows faster XXXXX times', ['image search', 'disparity range'])\n", "('NONE', 'the results show that a medium depth ( XXXXX of 4.7–7.0arc min ) makes the XXXXX slower', ['image search', 'disparity range'])\n", "('method used for task', 'increasing XXXXX increases channel resistance and XXXXX', ['contact resistance', 'o2 flow'])\n", "('NONE', 'visible decline in μfe as shorter XXXXX of devices with higher XXXXX', ['channel length', 'o2 flow'])\n", "('NONE', 'XXXXX results in reduced XXXXX in all ages', ['visual performance', 'discomfort glare'])\n", "('method used for task', 'a large XXXXX for building mems devices and for XXXXX', ['surface area', 'electrostatic interaction'])\n", "('method used for task', 'study of emissive XXXXX dependent recombination zone and XXXXX', ['energy transfer', 'layer layout'])\n", "('method used for task', 'study of emissive layer layout dependent XXXXX and XXXXX', ['energy transfer', 'recombination zone'])\n", "('NONE', 'study of emissive XXXXX dependent XXXXX and energy transfer', ['layer layout', 'recombination zone'])\n", "('NONE', 'XXXXX indicate that this phosphor also contains fast decay and slow XXXXX', ['decay graph', 'decay process'])\n", "('method used for task', 'XXXXX is easy to implement and XXXXX can reach to 99.3 %', ['motion detection', 'detection accuracy'])\n", "('method used for task', 'the moiré phenomenon is quantified using the XXXXX and XXXXX', ['contrast ratio', 'standard deviation'])\n", "('NONE', 'the effect of XXXXX on visually induced XXXXX ( vims ) and postural control was studied', ['visual motion', 'motion sickness'])\n", "('method used for task', 'visually induced XXXXX ( vims ) seems to be caused by XXXXX', ['motion sickness', 'visual motion'])\n", "('NONE', 'XXXXX indicate that this phosphor also contains fast decay and slow XXXXX', ['decay graph', 'decay process'])\n", "('NONE', 'the XXXXX of paper–pencil test was lower than tablet XXXXX', ['visual fatigue', 'pc test'])\n", "('method used for task', 'defining two terms of computer icons : XXXXX and XXXXX', ['action icon', 'knowledge icon'])\n", "('method used for task', 'analyzing the applied fields of XXXXX and XXXXX', ['action icon', 'knowledge icon'])\n", "('method used for task', 'factors from XXXXX , disparity energy and XXXXX are considered', ['spatial frequency', 'visual attention'])\n", "('method used for task', 'XXXXX is used for XXXXX', ['data embedding', 'sudoku technique'])\n", "('NONE', 'a reversible data hiding scheme that is based on the XXXXX and can achieve the higher XXXXX', ['embedding capacity', 'sudoku technique'])\n", "('method used for task', 'we suggest a new reversible data hiding algorithm based on the probabilistic XXXXX for XXXXX', ['secret sharing scheme', 'color images'])\n", "('NONE', 'the XXXXX and the task-completion time are inversely related . using a smaller display results in a shorter XXXXX', ['completion time', 'display size'])\n", "('NONE', 'effects of the XXXXX on the propagation of XXXXX have been investigated', ['system parameters', 'pressure waves'])\n", "('NONE', 'a facial one-pot XXXXX at XXXXX for ag2se nanoparticles has been developed', ['low temperature', 'synthesis method'])\n", "('method used for task', 'XXXXX ( field equations coupled with optical gain ) is solved by XXXXX using gpgpu card to reduce computation time and to increase accuracy by reducing integration step', ['numerical model', 'parallel computing'])\n", "('method used for task', 'XXXXX ( field equations coupled with optical gain ) is solved by parallel computing using gpgpu card to reduce XXXXX and to increase accuracy by reducing integration step', ['numerical model', 'computation time'])\n", "('method used for task', 'numerical model ( field equations coupled with optical gain ) is solved by XXXXX using gpgpu card to reduce XXXXX and to increase accuracy by reducing integration step', ['parallel computing', 'computation time'])\n", "('NONE', 'effective XXXXX in china need care about review format and XXXXX', ['voting system', 'review systems'])\n", "('NONE', 'we examine the XXXXX of XXXXX', ['moderating effect', 'product category'])\n", "('NONE', 'XXXXX capability plays a mediating role through which business achieves XXXXX', ['competitive advantage', 'business integration'])\n", "('NONE', 'we provided a XXXXX to illustrate the applicability of our XXXXX and our architecture in real life', ['case study', 'trust model'])\n", "('NONE', 'XXXXX and experimental results have shown the efficacy of the proposed XXXXX', ['theoretical analysis', 'incentive mechanism'])\n", "('method used for task', 'relationships between XXXXX and a XXXXX are moderated by negative valence of voc', ['customer expectations', 'brand community interaction'])\n", "('method used for task', 'the positive relationship between XXXXX and XXXXX is moderated by independent self-construal', ['customer satisfaction', 'customer loyalty'])\n", "('method used for task', 'we developed a XXXXX to depict the intention to self-disclose XXXXX in social commerce environment', ['personal information', 'research model'])\n", "('method used for task', 'the XXXXX relies on literature on XXXXX and communication privacy management theory', ['research model', 'information disclosure intention'])\n", "('method used for task', 'the XXXXX relies on literature on information disclosure intention and XXXXX management theory', ['research model', 'communication privacy'])\n", "('method used for task', 'the research model relies on literature on XXXXX and XXXXX management theory', ['information disclosure intention', 'communication privacy'])\n", "('method used for task', 'perceived enjoyment , perceived apathy and XXXXX positively affects intention to disclose XXXXX', ['perceived usefulness', 'personal information'])\n", "('NONE', 'fairness of XXXXX during XXXXX plays an important role in information disclosure intention', ['information exchange', 'social commerce'])\n", "('NONE', 'fairness of XXXXX during social commerce plays an important role in XXXXX', ['information exchange', 'information disclosure intention'])\n", "('NONE', 'fairness of information exchange during XXXXX plays an important role in XXXXX', ['social commerce', 'information disclosure intention'])\n", "('NONE', 'an inferior XXXXX should introduce a q & a service when the XXXXX difference is small', ['search engine', 'search quality'])\n", "('NONE', 'an inferior XXXXX should improve its XXXXX when the XXXXX difference is large', ['search engine', 'search quality'])\n", "('NONE', 'an inferior XXXXX should improve its XXXXX when the XXXXX difference is large', ['search engine', 'search quality'])\n", "('NONE', 'an inferior search engine should improve its XXXXX when the XXXXX difference is large', ['search quality', 'search difference'])\n", "('NONE', 'an inferior search engine should improve its XXXXX when the XXXXX difference is large', ['search quality', 'search difference'])\n", "('NONE', 'XXXXX confirms the effectiveness and value of our XXXXX', ['user study', 'reputation mechanism'])\n", "('method used for task', 'the relationship between XXXXX and XXXXX is moderated by impulse buying tendencies', ['product involvement', 'purchase intention'])\n", "('NONE', 'XXXXX requires XXXXX between banks and telecom operators', ['mobile payment', 'collective action'])\n", "('method used for task', 'social XXXXX and organizational support theory are used to explain XXXXX', ['knowledge contribution', 'exchange theory'])\n", "('method used for task', 'perceived XXXXX and perceived leader support positively affect XXXXX', ['knowledge contribution', 'community support'])\n", "('NONE', 'perceived community support and perceived XXXXX positively affect XXXXX', ['knowledge contribution', 'leader support'])\n", "('method used for task', 'perceived XXXXX and perceived XXXXX positively affect knowledge contribution', ['community support', 'leader support'])\n", "('method used for task', 'we identify the mediation effects of perceived XXXXX and perceived XXXXX', ['community support', 'leader support'])\n", "('NONE', 'XXXXX of reputation in on line transactions is presented in terms of XXXXX', ['theoretical analysis', 'game theory'])\n", "('NONE', 'peripheral cue moderates elaboration on XXXXX but not XXXXX', ['eye movements', 'purchase intention'])\n", "('method used for task', 'the relationship between XXXXX and XXXXX is more significant in high elaboration with negative cue and in low elaboration with positive cue', ['purchase intention', 'eye movement'])\n", "('NONE', 'this study examined drivers of interdependence and XXXXX in XXXXX in virtual communities', ['network convergence', 'social networks'])\n", "('NONE', 'validating the framework by a XXXXX of technology-based XXXXX in mobile payment ecosystem', ['case study', 'market collusion'])\n", "('method used for task', 'use mapreduce to adapt the serial XXXXX for XXXXX', ['map-matching algorithm', 'cloud computing environment'])\n", "('method used for task', 'ogb managers can leverage platform to enhance XXXXX and XXXXX', ['social capital', 'consumer benefits'])\n", "('NONE', 'active participation mediates the effect of XXXXX on ogb XXXXX', ['social capital', 'consumer benefits'])\n", "('method used for task', 'application of XXXXX to examine substitution effects in XXXXX', ['regression analysis', 'fashion e-commerce'])\n", "('NONE', 'XXXXX of multi-criteria cf XXXXX is improved', ['predictive accuracy', 'recommender systems'])\n", "('method used for task', 'XXXXX is affected by the XXXXX : search/purchase goods', ['channel choice', 'product category'])\n", "('NONE', 'incorporated XXXXX of the topic sentence to the XXXXX of sentiment classification', ['feature set', 'feature space'])\n", "('NONE', 'incorporated XXXXX of the topic sentence to the feature space of XXXXX', ['feature set', 'sentiment classification'])\n", "('NONE', 'incorporated feature set of the topic sentence to the XXXXX of XXXXX', ['feature space', 'sentiment classification'])\n", "('method used for task', 'the paper points out the XXXXX such as developing appropriate standardized service description and registry and proposing new ai based methods to achieve automated generation of service description for XXXXX , selection and composition', ['research directions', 'service discovery'])\n", "('NONE', 'the paper points out the XXXXX such as developing appropriate standardized XXXXX and registry and proposing new ai based methods to achieve automated generation of XXXXX for service discovery , selection and composition', ['research directions', 'service description'])\n", "('NONE', 'the paper points out the XXXXX such as developing appropriate standardized XXXXX and registry and proposing new ai based methods to achieve automated generation of XXXXX for service discovery , selection and composition', ['research directions', 'service description'])\n", "('method used for task', 'the paper points out the research directions such as developing appropriate standardized XXXXX and registry and proposing new ai based methods to achieve automated generation of XXXXX for XXXXX , selection and composition', ['service discovery', 'service description'])\n", "('method used for task', 'the paper points out the research directions such as developing appropriate standardized XXXXX and registry and proposing new ai based methods to achieve automated generation of XXXXX for XXXXX , selection and composition', ['service discovery', 'service description'])\n", "('NONE', 'trust can mediate the effects of perceived personalization and XXXXX on XXXXX', ['privacy concerns', 'acceptance intention'])\n", "('NONE', 'for elder potential users , XXXXX have no influences on XXXXX', ['privacy concerns', 'acceptance intention'])\n", "('NONE', 'empirical discovery of the XXXXX of group-level XXXXX', ['mixed effects', 'social capital'])\n", "('method used for task', 'XXXXX and experience affect XXXXX and intentions', ['tie strength', 'message credibility'])\n", "('method used for task', 'we model users’ mobile XXXXX based on a multi-state XXXXX', ['online behavior', 'markov model'])\n", "('method used for task', 'we model users’ mobile XXXXX based on a XXXXX', ['online behavior', 'hidden markov model'])\n", "('NONE', 'typical XXXXX of mobile XXXXX are discovered and analyzed', ['sequential patterns', 'online behavior'])\n", "('method used for task', 'we use bayesian XXXXX to estimate the XXXXX of qoe , considering available prior quality of service ( qos ) data and user ratings', ['data analysis', 'posterior distribution'])\n", "('NONE', 'this is the first study examining XXXXX in XXXXX', ['gender differences', 'online auctions'])\n", "('NONE', 'males exhibit a higher level of XXXXX in XXXXX', ['social interaction', 'online auctions'])\n", "('method used for task', 'we studied the effect of edm and pmedm parameters on XXXXX , wlt and XXXXX of d2 steel', ['fatigue life', 'heat flux'])\n", "('method used for task', 'analyzing and developing XXXXX for the total XXXXX generated by using fem', ['numerical models', 'heat flux'])\n", "('NONE', 'a method for XXXXX of XXXXX is presented', ['automatic generation', 'water distribution systems'])\n", "('NONE', 'XXXXX of lake and XXXXX is generally inadequate', ['performance assessment', 'marine models'])\n", "('NONE', 'designed to allow flexibility in the approach to construct different XXXXX without compiling XXXXX', ['crop models', 'source code'])\n", "('NONE', 'location tracking through time of XXXXX output for ease of XXXXX', ['simulation model', 'decision making'])\n", "('method used for task', 'we review XXXXX related technologies to manage , transfer and process XXXXX', ['web service', 'big data'])\n", "('method used for task', 'a novel muscl scheme is developed to provide more efficient and accurate XXXXX to the 2d XXXXX', ['numerical solutions', 'shallow water equations'])\n", "('NONE', 'XXXXX showed that increasing tree height and local XXXXX were the main factors associated with damage', ['statistical analysis', 'wind speed'])\n", "('NONE', 'XXXXX automatically scripts manual XXXXX data edits in python', ['quality control', 'odm tools'])\n", "('NONE', 'XXXXX preserves provenance of XXXXX edits', ['quality control', 'odm tools'])\n", "('method used for task', 'XXXXX is XXXXX and cross platform compatible', ['open source', 'odm tools'])\n", "('method used for task', 'uncertainties on XXXXX and XXXXX prevail after 2080', ['sea-level rise', 'climate change scenarios'])\n", "('NONE', 'effectiveness of XXXXX depends on global XXXXX', ['boundary conditions', 'policy strategies'])\n", "('NONE', 'we present an overview of sa and its link to XXXXX , XXXXX and evaluation , robust decision-making', ['uncertainty analysis', 'model calibration'])\n", "('NONE', 'XXXXX was assessed through a XXXXX', ['network structure', 'participatory approach'])\n", "('NONE', 'XXXXX improved XXXXX and model response', ['data integration', 'data quality'])\n", "('NONE', 'XXXXX improved data quality and XXXXX', ['data integration', 'model response'])\n", "('method used for task', 'data integration improved XXXXX and XXXXX', ['data quality', 'model response'])\n", "('method used for task', 'application of XXXXX to co-construct XXXXX of flood risk', ['bayesian networks', 'conceptual model'])\n", "('NONE', 'we developed a model for the XXXXX of XXXXX and maintenance', ['joint optimization', 'process control'])\n", "('NONE', 'potential XXXXX can be obtained from joint spc and XXXXX', ['cost savings', 'maintenance policies'])\n", "('method used for task', 'we discuss their XXXXX and present suitable XXXXX', ['computational complexity', 'solution procedures'])\n", "('method used for task', 'we show the basic XXXXX to be solvable in XXXXX', ['allocation problem', 'polynomial time'])\n", "('NONE', 'we obtain a surrogate duality theorems for semi-definite XXXXX in the face of XXXXX', ['optimization problems', 'data uncertainty'])\n", "('method used for task', 'revise comparative properties of the XXXXX for the XXXXX', ['continuous model', 'discrete model'])\n", "('method used for task', 'the objective of this paper is the research on optimization methodologies for a complete integration of XXXXX , crew scheduling and driver rostering problems so as to develop XXXXX', ['vehicle scheduling', 'efficient algorithms'])\n", "('method used for task', 'we have developed a XXXXX based on XXXXX to solve the problem', ['heuristic approach', 'benders decomposition'])\n", "('NONE', 'results are reported on new XXXXX , derived from well-known XXXXX', ['routing problems', 'benchmark instances'])\n", "('method used for task', 'we develop an XXXXX to solve clusterwise XXXXX', ['incremental algorithm', 'linear regression problems'])\n", "('method used for task', 'we establish a formal link between XXXXX and spectral XXXXX', ['expected utility', 'risk measures'])\n", "('NONE', 'we consider three XXXXX applied in XXXXX', ['portfolio theory', 'quadratic optimization problems'])\n", "('NONE', 'the dm makes XXXXX of identified pareto optimal points and gives XXXXX', ['pairwise comparisons', 'reference points'])\n", "('method used for task', 'we propose XXXXX for the XXXXX of a speciality oils’ company', ['linear models', 'supply chain'])\n", "('method used for task', 'this article provides an XXXXX for a serial XXXXX with multiple suppliers and time varying demand', ['inventory model', 'supply chain'])\n", "('NONE', 'our results show that XXXXX and holding costs can affect XXXXX and order lot sizing decisions', ['supplier selection', 'inventory setup'])\n", "('NONE', 'involve XXXXX , XXXXX and bootstrapping', ['discrete-event simulation', 'robust optimization'])\n", "('NONE', 'to explore XXXXX under XXXXX with consumer returns', ['inventory control problem', 'consignment contract'])\n", "('NONE', 'to explore XXXXX under consignment contract with XXXXX', ['inventory control problem', 'consumer returns'])\n", "('NONE', 'to explore inventory control problem under XXXXX with XXXXX', ['consignment contract', 'consumer returns'])\n", "('method used for task', 'the main findings of the XXXXX is that XXXXX increases the economic order quantity and could serve as a buyer–supplier coordination mechanism', ['literature review', 'trade credit'])\n", "('method used for task', 'the main findings of the XXXXX is that trade credit increases the economic order quantity and could serve as a buyer–supplier XXXXX', ['literature review', 'coordination mechanism'])\n", "('method used for task', 'the main findings of the XXXXX is that trade credit increases the economic XXXXX and could serve as a buyer–supplier coordination mechanism', ['literature review', 'order quantity'])\n", "('NONE', 'the main findings of the literature review is that XXXXX increases the economic order quantity and could serve as a buyer–supplier XXXXX', ['trade credit', 'coordination mechanism'])\n", "('NONE', 'the main findings of the literature review is that XXXXX increases the economic XXXXX and could serve as a buyer–supplier coordination mechanism', ['trade credit', 'order quantity'])\n", "('NONE', 'the main findings of the literature review is that trade credit increases the economic XXXXX and could serve as a buyer–supplier XXXXX', ['coordination mechanism', 'order quantity'])\n", "('NONE', 'we derive a detailed agenda for XXXXX around two core themes : opportunities arising from inside XXXXX and opportunities arising from outside XXXXX', ['future research', 'operations management'])\n", "('NONE', 'we derive a detailed agenda for XXXXX around two core themes : opportunities arising from inside XXXXX and opportunities arising from outside XXXXX', ['future research', 'operations management'])\n", "('NONE', 'cooperation and XXXXX improve ecologic as well as XXXXX', ['vertical integration', 'economic efficiency'])\n", "('NONE', 'we model the XXXXX that deals with products having high XXXXX', ['supply chain', 'demand uncertainty'])\n", "('method used for task', 'we develop the bidirectional XXXXX to coordinate the XXXXX', ['supply chain', 'option contracts'])\n", "('NONE', 'additional XXXXX as well as XXXXX are proposed to significantly speed up overall solution time', ['lower bounds', 'valid inequalities'])\n", "('NONE', 'additional XXXXX as well as valid inequalities are proposed to significantly speed up overall XXXXX', ['lower bounds', 'solution time'])\n", "('method used for task', 'additional lower bounds as well as XXXXX are proposed to significantly speed up overall XXXXX', ['valid inequalities', 'solution time'])\n", "('method used for task', 'proposed an XXXXX for the two- and three-echelon reverse XXXXX', ['analytical model', 'supply chain'])\n", "('method used for task', 'defined the pc recycling XXXXX as foundation for generating the XXXXX', ['supply chain', 'analytical model'])\n", "('NONE', 'the XXXXX of our model is not only robust but also has a desired XXXXX', ['optimal portfolio', 'factor exposure'])\n", "('NONE', 'interrelationship between XXXXX , XXXXX , and ghgs is considered', ['lead time', 'transportation cost'])\n", "('method used for task', 'a novel solution heuristic for the general lotsizing and XXXXX for parallel XXXXX is presented', ['scheduling problem', 'production lines'])\n", "('NONE', 'XXXXX are calculated by an algorithmic approach using XXXXX', ['nash equilibria', 'linear programming'])\n", "('NONE', 'XXXXX increases logarithmically with number of discrete points , XXXXX decreases linearly', ['problem size', 'optimality gap'])\n", "('NONE', 'the above ideas are demonstrated with a XXXXX real world XXXXX', ['large scale', 'case study'])\n", "('method used for task', 'a XXXXX is suggested to represent the new XXXXX', ['mathematical programming model', 'maintenance scheduling problem'])\n", "('NONE', 'to show the applicability of the model and XXXXX , a XXXXX is reported on real data', ['case study', 'solution algorithm'])\n", "('method used for task', 'XXXXX for block-angular problems based on XXXXX', ['interior-point method', 'hybrid approach'])\n", "('NONE', 'XXXXX successfully imputes XXXXX in time series', ['variable neighborhood search', 'missing values'])\n", "('NONE', 'XXXXX successfully imputes missing values in XXXXX', ['variable neighborhood search', 'time series'])\n", "('NONE', 'variable neighborhood search successfully imputes XXXXX in XXXXX', ['missing values', 'time series'])\n", "('method used for task', 'a heuristic XXXXX based on the XXXXX and subgradient methods is provided', ['lagrangian relaxation', 'solution approach'])\n", "('method used for task', 'we solve the XXXXX in short cpu times to XXXXX below 5 %', ['problem instances', 'optimality gaps'])\n", "('method used for task', 'we develop a numerical nonlinear principal XXXXX ( not tractable to solve in XXXXX )', ['closed form', 'agent problem'])\n", "('method used for task', 'XXXXX for XXXXX of markowitz portfolio weights', ['closed-form formulas', 'estimation errors'])\n", "('NONE', 'XXXXX for estimation errors of XXXXX weights', ['closed-form formulas', 'markowitz portfolio'])\n", "('NONE', 'closed-form formulas for XXXXX of XXXXX weights', ['estimation errors', 'markowitz portfolio'])\n", "('NONE', 'efficient bootstrap-based method for XXXXX of XXXXX', ['empirical assessment', 'estimation errors'])\n", "('NONE', 'we identified five XXXXX that close XXXXX . then we gave a review of each model', ['research streams', 'surrogate duality gaps'])\n", "('method used for task', 'multiperiod XXXXX ( msl ) policies for XXXXX', ['service level', 'supply chains'])\n", "('NONE', 'providing XXXXX of decision variables of specific XXXXX', ['explicit solution', 'dynamic models'])\n", "('NONE', 'providing XXXXX of XXXXX of specific dynamic models', ['explicit solution', 'decision variables'])\n", "('NONE', 'providing explicit solution of XXXXX of specific XXXXX', ['dynamic models', 'decision variables'])\n", "('method used for task', 'finding relationship between the dynamics of XXXXX and XXXXX', ['improvement activities', 'model structures'])\n", "('NONE', 'we introduce the first modeling as full truckload pickup and XXXXX with XXXXX', ['time windows', 'delivery problem'])\n", "('NONE', 'a XXXXX with permutation chromosome and XXXXX is proposed', ['genetic algorithm', 'local search'])\n", "('NONE', 'information granulation of XXXXX used in XXXXX', ['linguistic information', 'group decision making'])\n", "('method used for task', 'XXXXX is used to made operational the XXXXX', ['granular computing', 'linguistic information'])\n", "('NONE', 'XXXXX expressed in terms of XXXXX defined as sets', ['linguistic information', 'information granules'])\n", "('method used for task', 'the granulation of the XXXXX is formulated as an XXXXX', ['linguistic terms', 'optimization problem'])\n", "('method used for task', 'XXXXX need to be readjusted to account for XXXXX', ['risk measures', 'default risk'])\n", "('NONE', 'XXXXX compare against the best known XXXXX', ['computational experiments', 'exact solution methods'])\n", "('NONE', 'XXXXX using XXXXX for packing circles is proposed', ['heuristic algorithm', 'linear models'])\n", "('method used for task', 'an inverse mean-deviation XXXXX finds a deviation measure for a given XXXXX', ['optimal portfolio', 'portfolio problem'])\n", "('method used for task', 'an inverse mean-deviation portfolio problem finds a XXXXX for a given XXXXX', ['optimal portfolio', 'deviation measure'])\n", "('NONE', 'an inverse mean-deviation XXXXX finds a XXXXX for a given optimal portfolio', ['portfolio problem', 'deviation measure'])\n", "('NONE', 'necessary and XXXXX for the existence of such a XXXXX are obtained', ['sufficient conditions', 'deviation measure'])\n", "('NONE', 'we consider two-stage XXXXX with arc XXXXX', ['vehicle routing problem', 'time windows'])\n", "('method used for task', 'we propose a mip formulation and a XXXXX based on XXXXX', ['heuristic approach', 'memetic algorithm'])\n", "('method used for task', 'we propose a XXXXX and a XXXXX based on memetic algorithm', ['heuristic approach', 'mip formulation'])\n", "('method used for task', 'we propose a XXXXX and a heuristic approach based on XXXXX', ['memetic algorithm', 'mip formulation'])\n", "('NONE', 'the XXXXX obtains the XXXXX of small and medium-size problems', ['optimal solution', 'mip formulation'])\n", "('NONE', 'the XXXXX obtains good quality solutions in a short XXXXX', ['memetic algorithm', 'computation time'])\n", "('NONE', 'the XXXXX obtains good XXXXX in a short computation time', ['memetic algorithm', 'quality solutions'])\n", "('NONE', 'the memetic algorithm obtains good XXXXX in a short XXXXX', ['computation time', 'quality solutions'])\n", "('NONE', 'XXXXX , under conditions of XXXXX , is reviewed', ['workforce planning', 'market uncertainty'])\n", "('NONE', 'decisions regarding XXXXX , targeting different XXXXX , are discussed', ['service levels', 'workforce shifts'])\n", "('method used for task', 'develop XXXXX to construct tdsvm XXXXX', ['memetic algorithm', 'classification models'])\n", "('method used for task', 'we use XXXXX to characterize the XXXXX', ['stochastic optimal control', 'optimal policies'])\n", "('NONE', 'XXXXX , XXXXX , input requirement functions are considered', ['production function', 'distance functions'])\n", "('method used for task', 'provides a XXXXX and a decomposition-based meta-heuristic XXXXX', ['model formulation', 'solution method'])\n", "('method used for task', 'proposed XXXXX and a hybrid heuristic solution technique for solving a XXXXX variant', ['mathematical model', 'vehicle routing problem'])\n", "('method used for task', 'proposed XXXXX and a hybrid heuristic XXXXX for solving a vehicle routing problem variant', ['mathematical model', 'solution technique'])\n", "('method used for task', 'proposed mathematical model and a hybrid heuristic XXXXX for solving a XXXXX variant', ['vehicle routing problem', 'solution technique'])\n", "('NONE', 'we study a XXXXX with soft XXXXX and stochastic travel times', ['vehicle routing problem', 'time windows'])\n", "('method used for task', 'we propose an XXXXX based on XXXXX and branch-and-price', ['column generation', 'exact solution approach'])\n", "('NONE', 'a new formulation for the time-dependent multi-zone multi-trip XXXXX with XXXXX', ['vehicle routing problem', 'time window'])\n", "('NONE', 'we introduce a law of one price rule for XXXXX of XXXXX', ['shadow prices', 'undesirable outputs'])\n", "('method used for task', 'a near-optimal joint XXXXX is found that considers XXXXX and batching', ['stochastic demand', 'inventory policy'])\n", "('NONE', 'we study the XXXXX with sequence-independent XXXXX', ['setup times', 'two-machine flowshop problem'])\n", "('method used for task', 'we develop an XXXXX for serial XXXXX', ['supply chains', 'optimization procedure'])\n", "('method used for task', 'discrete malliavin XXXXX is introduced to compute greeks ( sensitivity ) of the XXXXX', ['calculus approach', 'option prices'])\n", "('NONE', 'achieve better XXXXX than a commercial stock cutting XXXXX', ['software package', 'solution quality'])\n", "('method used for task', 'control strategies include rts ( regular XXXXX ) assignment and XXXXX', ['time slots', 'rts reservation'])\n", "('method used for task', 'XXXXX improves the unused time slots and XXXXX of rts assignment', ['waiting times', 'rts reservation'])\n", "('method used for task', 'rts reservation improves the unused XXXXX and XXXXX of rts assignment', ['waiting times', 'time slots'])\n", "('NONE', 'XXXXX improves the unused XXXXX and waiting times of rts assignment', ['rts reservation', 'time slots'])\n", "('NONE', 'XXXXX decreases the XXXXX under the cvar', ['capacity uncertainty', 'order quantity'])\n", "('NONE', 'XXXXX may increase or decrease the XXXXX under the var constraint', ['capacity uncertainty', 'order quantity'])\n", "('NONE', 'XXXXX may increase or decrease the order quantity under the XXXXX', ['capacity uncertainty', 'var constraint'])\n", "('NONE', 'capacity uncertainty may increase or decrease the XXXXX under the XXXXX', ['order quantity', 'var constraint'])\n", "('method used for task', 'under the XXXXX , the effect depends on the XXXXX', ['confidence level', 'var constraint'])\n", "('method used for task', 'conduct the XXXXX for the XXXXX', ['time complexity analysis', 'reduction procedures'])\n", "('method used for task', 'XXXXX and XXXXX illustrate the usefulness of theoretical results', ['numerical examples', 'case study'])\n", "('NONE', 'we measure XXXXX through an output-oriented weighted XXXXX', ['additive model', 'revenue inefficiency'])\n", "('NONE', 'the XXXXX can be found by a XXXXX', ['dynamic programming algorithm', 'optimal routing strategy'])\n", "('NONE', 'the XXXXX has a specific XXXXX when the time horizon is finite or infinite', ['optimal strategy', 'threshold structure'])\n", "('NONE', 'the XXXXX has a specific threshold structure when the XXXXX is finite or infinite', ['optimal strategy', 'time horizon'])\n", "('NONE', 'the optimal strategy has a specific XXXXX when the XXXXX is finite or infinite', ['threshold structure', 'time horizon'])\n", "('method used for task', 'the joint XXXXX is characterized in a continuous XXXXX', ['optimal policy', 'inventory review'])\n", "('NONE', 'considers fractionated intensity-modulated XXXXX of XXXXX under breathing motion uncertainty', ['radiation therapy', 'lung cancer'])\n", "('method used for task', 'proposes a method that generalizes XXXXX and adaptive XXXXX', ['robust optimization', 'radiation therapy'])\n", "('NONE', 'isotonic nonparametric XXXXX ( inls ) fits a XXXXX to data', ['least squares', 'step function'])\n", "('NONE', 'free disposal hull ( fdh ) envelops all XXXXX by XXXXX', ['data points', 'step function'])\n", "('method used for task', 'combined XXXXX , meta-heuristic and XXXXX', ['local search', 'mathematical programming'])\n", "('NONE', 'XXXXX of parameters in the XXXXX', ['stress testing', 'markowitz model'])\n", "('method used for task', 'simultaneous XXXXX and XXXXX is formulated in a job shop', ['lot sizing', 'scheduling problem'])\n", "('method used for task', 'simultaneous XXXXX and scheduling problem is formulated in a XXXXX', ['lot sizing', 'job shop'])\n", "('method used for task', 'simultaneous lot sizing and XXXXX is formulated in a XXXXX', ['scheduling problem', 'job shop'])\n", "('method used for task', 'XXXXX , buyer–supplier relationships and XXXXX are considered', ['knowledge sharing', 'production capacities'])\n", "('method used for task', 'negative correlation , although more erratic , improves XXXXX and XXXXX relative to iid', ['lead time', 'inventory performance'])\n", "('NONE', 'conducts a XXXXX to assess the accuracy of the XXXXX', ['simulation study', 'scenario model'])\n", "('method used for task', 'a mixed XXXXX is introduced for a real short sea XXXXX', ['inventory routing problem', 'integer model'])\n", "('method used for task', 'the intm–intn XXXXX is based on XXXXX', ['decision making', 'choice value soft sets'])\n", "('method used for task', 'computing XXXXX is a heavy load when processing XXXXX', ['big data', 'choice value soft sets'])\n", "('NONE', 'we develop a XXXXX without computation of XXXXX', ['pruning method', 'choice value soft sets'])\n", "('NONE', 'we conclude our XXXXX with XXXXX directions', ['future research', 'review paper'])\n", "('method used for task', 'the XXXXX is decomposed into stage efficiencies in XXXXX', ['system efficiency', 'product form'])\n", "('method used for task', 'develops the XXXXX approach to XXXXX', ['inventory balance', 'order forecasting'])\n", "('NONE', 'XXXXX approach outperforms traditional XXXXX approaches', ['inventory balance', 'order forecasting'])\n", "('NONE', 'we discuss in detail an application in the XXXXX of XXXXX', ['strategic planning', 'freight transportation'])\n", "('method used for task', 'XXXXX and XXXXX for electricity generation is analyzed', ['technology selection', 'capacity investment'])\n", "('NONE', 'synthesis of unique context of XXXXX within a coherent XXXXX framework', ['product families', 'equilibrium decision'])\n", "('NONE', 'we present XXXXX with efficient XXXXX in each iteration', ['tabu search algorithm', 'improvement method'])\n", "('method used for task', 'lagragian XXXXX is proposed to obtain the XXXXX', ['relaxation algorithm', 'lower bounds'])\n", "('method used for task', 'it shows that results of XXXXX are very close to the XXXXX', ['tabu search', 'lower bounds'])\n", "('method used for task', 'a generalized multiplicative XXXXX ( gmddf ) is developed to measure XXXXX ( te )', ['directional distance function', 'technical efficiency'])\n", "('NONE', 'quantifies the effectiveness of a firm’s XXXXX under XXXXX', ['production system', 'demand uncertainty'])\n", "('NONE', 'we assess trade-offs between socio-economic and XXXXX of XXXXX', ['environmental impact', 'manure management'])\n", "('method used for task', 'we apply XXXXX to select best compromise XXXXX solutions', ['compromise programming', 'manure management'])\n", "('NONE', 'our procedures generate strong XXXXX and improve the convergence of XXXXX', ['lower bounds', 'branch-and-bound algorithms'])\n", "('method used for task', 'XXXXX and other XXXXX ( psms ) can help address science and technology risk conflicts', ['problem structuring methods', 'issues mapping'])\n", "('method used for task', 'we present a XXXXX to solve the combined XXXXX', ['branch-and-price algorithm', 'optimization problem'])\n", "('NONE', 'show impact of XXXXX on XXXXX', ['capacity constraints', 'optimal policies'])\n", "('NONE', 'ignoring the XXXXX in XXXXX may lead to inadequate decisions', ['supplier selection', 'currency uncertainty'])\n", "('NONE', 'XXXXX on XXXXX showing that significant improvements can be obtained', ['computational study', 'real-world data'])\n", "('method used for task', 'we derive XXXXX for the ( s−1 , s ) XXXXX with backordering', ['structural properties', 'inventory model'])\n", "('NONE', 'XXXXX without making a XXXXX', ['column generation', 'linear programming relaxation'])\n", "('method used for task', 'customised XXXXX to determine when the restricted master problem contains an XXXXX', ['optimality conditions', 'optimal solution'])\n", "('method used for task', 'customised XXXXX to determine when the restricted XXXXX contains an optimal solution', ['optimality conditions', 'master problem'])\n", "('NONE', 'customised optimality conditions to determine when the restricted XXXXX contains an XXXXX', ['optimal solution', 'master problem'])\n", "('NONE', 'develop equilibrium variety and XXXXX strategies in a retailer-led XXXXX', ['supply chain', 'channel structure'])\n", "('method used for task', 'comparisons based on XXXXX and XXXXX', ['objective function value', 'time consumption'])\n", "('method used for task', 'we build XXXXX for XXXXX in foreclosed housing acquisition', ['stochastic models', 'budget allocation'])\n", "('method used for task', 'interactive XXXXX for handling many objective XXXXX', ['evolutionary algorithm', 'optimization problems'])\n", "('method used for task', 'a XXXXX for XXXXX in copper mines is presented', ['stochastic model', 'strategic planning'])\n", "('NONE', 'XXXXX can enable XXXXX in engineering organizations', ['action research', 'technology management'])\n", "('method used for task', 'a new velocity updating strategy is incorporated into XXXXX to improve XXXXX', ['pso algorithm', 'solution quality'])\n", "('NONE', 'depending on the XXXXX , total XXXXX can become negative', ['model parameters', 'agency costs'])\n", "('method used for task', 'we propose a XXXXX for XXXXX', ['evolutionary multiobjective optimization', 'resource allocation model'])\n", "('method used for task', 'we propose a double-sphere XXXXX for XXXXX', ['crowding distance', 'evolutionary multiobjective optimization'])\n", "('method used for task', 'experiments are performed on twelve XXXXX and one XXXXX', ['benchmark problems', 'real-world problem'])\n", "('method used for task', 'XXXXX and evolutionary based XXXXX are implemented', ['local search', 'solution algorithms'])\n", "('method used for task', 'designed a customized XXXXX to solve XXXXX of practical size', ['memetic algorithm', 'problem instances'])\n", "('method used for task', 'propose a XXXXX and XXXXX for the problem', ['mixed integer programming', 'branch-and-bound algorithm'])\n", "('method used for task', 'we build a corresponding XXXXX for XXXXX and project selection at the level of software skills', ['mathematical model', 'resource allocation'])\n", "('NONE', 'continuous-review XXXXX with XXXXX', ['time-varying parameters', 'inventory problem'])\n", "('NONE', 'classification of the XXXXX within the XXXXX', ['goal programming model', 'portfolio selection'])\n", "('method used for task', 'the perspectives of the XXXXX for the XXXXX', ['goal programming', 'portfolio selection'])\n", "('NONE', 'XXXXX for the competitiveness of the XXXXX', ['problem structuring', 'automotive industry'])\n", "('NONE', 'analysis of competitiveness of the XXXXX using bayesian XXXXX', ['automotive industry', 'causal networks'])\n", "('NONE', 'XXXXX using bayesian XXXXX', ['sensitivity analysis', 'causal networks'])\n", "('method used for task', 'uhgs matches or outperforms all current best problem-tailored algorithms on 29 notable XXXXX and 42 XXXXX', ['vrp variants', 'benchmark sets'])\n", "('NONE', 'can handle the large-scale XXXXX with fairly good XXXXX', ['problem instances', 'solution quality'])\n", "('method used for task', 'namely standard XXXXX and the XXXXX', ['quadratic programming', 'quadratic assignment problem'])\n", "('NONE', 'two-level XXXXX under continuous review with XXXXX', ['supply chain', 'uncertain demand'])\n", "('method used for task', 'a necessary and XXXXX is found for generating an XXXXX', ['sufficient condition', 'optimal solution'])\n", "('method used for task', 'sgp problems are difficult , XXXXX to solve for XXXXX', ['np-hard problems', 'global optimality'])\n", "('method used for task', 'an XXXXX for the cm to derive a reliable XXXXX from a pair-wise comparison matrix', ['optimization model', 'priority vector'])\n", "('method used for task', 'an XXXXX for the cm to derive a reliable priority vector from a pair-wise XXXXX', ['optimization model', 'comparison matrix'])\n", "('NONE', 'an optimization model for the cm to derive a reliable XXXXX from a pair-wise XXXXX', ['priority vector', 'comparison matrix'])\n", "('NONE', 'compare cm with other prioritization methods based on two XXXXX : XXXXX and minimum violation', ['performance evaluation', 'euclidean distance'])\n", "('method used for task', 'induce a XXXXX for a pair-wise XXXXX based on the cm', ['consistency index', 'comparison matrix'])\n", "('method used for task', 'the XXXXX are based on optimality arguments and XXXXX', ['linear programming', 'reduction procedures'])\n", "('NONE', 'XXXXX with three sequential mixed and integer XXXXX', ['optimization methods', 'linear programs'])\n", "('method used for task', 'we propose uacor , a unified XXXXX for XXXXX', ['continuous optimization', 'aco algorithm'])\n", "('NONE', 'we characterize the XXXXX with XXXXX of purchasing time', ['rational expectations equilibrium', 'customer choice'])\n", "('NONE', 'we show how information and XXXXX affect the XXXXX', ['demand dynamics', 'rational expectations equilibrium'])\n", "('NONE', 'found XXXXX oppose the logic of XXXXX', ['fuzzy numbers', 'fuzzy set theory'])\n", "('method used for task', 'found XXXXX can not give an acceptable method to rank XXXXX', ['fuzzy ahp', 'fuzzy numbers'])\n", "('method used for task', 'the heuristics are tested using real-world XXXXX and XXXXX where a congestion charge scheme is in operation', ['road networks', 'traffic data'])\n", "('NONE', 'present a XXXXX based XXXXX', ['lagrangian relaxation', 'solution method'])\n", "('method used for task', 'a XXXXX is proposed to obtain good lower and XXXXX of the problem', ['hybrid approach', 'upper bounds'])\n", "('method used for task', 'we propose XXXXX to improve the XXXXX of the proposed formulation', ['valid inequalities', 'lower bound'])\n", "('NONE', 'for some XXXXX , the XXXXX is a simple threshold policy', ['optimal policy', 'parameter values'])\n", "('method used for task', 'we apply XXXXX and dynamic XXXXX to solve the problem', ['column generation', 'constraint aggregation'])\n", "('NONE', 'we improve lower and XXXXX on existing XXXXX', ['upper bounds', 'benchmark instances'])\n", "('method used for task', 'minimizing the real XXXXX is introduced as a practical XXXXX', ['travel time', 'robustness measure'])\n", "('method used for task', 'we propose XXXXX and XXXXX to model service times', ['finite mixtures', 'dirichlet processes'])\n", "('method used for task', 'we propose an innovative XXXXX for clustering categorical XXXXX', ['hybrid method', 'sequential data'])\n", "('NONE', 'we propose two XXXXX which determine the players’ XXXXX', ['optimal strategies', 'solution procedures'])\n", "('method used for task', 'this paper uses a XXXXX to measure the XXXXX index', ['probabilistic analysis', 'malmquist productivity'])\n", "('NONE', 'we propose a new XXXXX ; it computes the XXXXX for small cases', ['global optimum', 'milp formulation'])\n", "('method used for task', 'the scc XXXXX is modeled as a XXXXX ( mip ) problem', ['scheduling problem', 'mixed-integer programming'])\n", "('method used for task', 'we propose simple and XXXXX to solve those XXXXX', ['efficient algorithms', 'optimization problems'])\n", "('method used for task', 'a strongly XXXXX is proposed to solve the XXXXX', ['polynomial-time algorithm', 'inverse problem'])\n", "('method used for task', 'propose a XXXXX based on sequential bifurcation and XXXXX', ['response surface methodology', 'solution framework'])\n", "('NONE', 'we analyze an XXXXX ( ev ) XXXXX under a price-discount scheme', ['electric vehicle', 'supply chain'])\n", "('method used for task', 'use of XXXXX to improve XXXXX', ['social welfare', 'cross-border system'])\n", "('method used for task', 'a new XXXXX for XXXXX is proposed and used', ['classification scheme', 'transport operations'])\n", "('NONE', 'we provide explicit XXXXX on the multiple as function of past XXXXX and volatilities', ['upper bounds', 'asset returns'])\n", "('method used for task', 'we show how this new XXXXX performs for various financial XXXXX', ['asset allocation method', 'market conditions'])\n", "('method used for task', 'insufficiency of XXXXX to estimate an XXXXX is identified', ['linear relaxation solutions', 'upper bound set'])\n", "('method used for task', 'a heuristic using XXXXX and a XXXXX is proposed', ['priority rules', 'column generation algorithm'])\n", "('NONE', 'a XXXXX solves the problem with varying XXXXX', ['resource availability', 'column generation algorithm'])\n", "('NONE', 'we realise an XXXXX for hedge funds using XXXXX approach', ['early warning system', 'regression trees'])\n", "('method used for task', 'XXXXX and equal XXXXX are its boundary members', ['shapley value', 'division value'])\n", "('NONE', 'numerical applications in XXXXX of XXXXX', ['portfolio optimization', 'stock markets'])\n", "('NONE', 'we compute an XXXXX with a least XXXXX heuristic', ['upper bound', 'discrepancy search'])\n", "('NONE', 'we consider a two-echelon XXXXX with poisson demand and XXXXX', ['lost sales', 'inventory system'])\n", "('method used for task', 'constant XXXXX for XXXXX with sigmoid utility', ['knapsack problem', 'factor algorithm'])\n", "('method used for task', 'constant XXXXX for XXXXX with sigmoid utility', ['generalized assignment problem', 'factor algorithm'])\n", "('NONE', 'it initially deals with XXXXX in possibilistic XXXXX', ['dynamic systems', 'optimal stopping problems'])\n", "('method used for task', 'the presence of a unique reorder level wherever the XXXXX is insensitive to certain XXXXX', ['order quantity', 'parameter values'])\n", "('NONE', 'this simple XXXXX can be further refined making use of all the short-term XXXXX available', ['priority rule', 'demand information'])\n", "('method used for task', 'an extensive XXXXX is presented on node XXXXX with intermediate facilities', ['literature review', 'routing problems'])\n", "('method used for task', 'four XXXXX for amblp are analyzed for XXXXX', ['heuristic methods', 'worst-case performance'])\n", "('NONE', 'we model a flexible XXXXX with n market periods and XXXXX', ['demand uncertainty', 'capacity strategy'])\n", "('NONE', 'locating XXXXX by conventional methods makes XXXXX vulnerable', ['facts devices', 'power systems'])\n", "('NONE', 'give a model for the XXXXX of the legislative council of XXXXX ( legco )', ['voting system', 'hong kong'])\n", "('NONE', 'an XXXXX of u.s. coal-fired XXXXX operating in 2011', ['empirical study', 'power plants'])\n", "('method used for task', 'the portfolio problem solved using XXXXX and XXXXX', ['dynamic programming', 'monte carlo methods'])\n", "('NONE', 'the XXXXX solved using XXXXX and monte carlo methods', ['dynamic programming', 'portfolio problem'])\n", "('method used for task', 'the XXXXX solved using dynamic programming and XXXXX', ['monte carlo methods', 'portfolio problem'])\n", "('NONE', 'minimization of the XXXXX of any desired XXXXX of the hedging error', ['expected value', 'penalty function'])\n", "('method used for task', 'innovative integration of XXXXX and the well-known ( s , s ) XXXXX', ['capacity strategy', 'inventory policy'])\n", "('method used for task', 'an improved kfp-mco classifier based kernel , fuzzification , and XXXXX is proposed and used for XXXXX', ['credit scoring', 'penalty factors'])\n", "('method used for task', 'we illustrate that parametric bayesian updates based on observed XXXXX can be used effectively for XXXXX', ['sales data', 'demand learning'])\n", "('method used for task', 'dynamic XXXXX is larger before the XXXXX than during it', ['financial crisis', 'cost inefficiency'])\n", "('NONE', 'introducing a new XXXXX : tollbooth XXXXX', ['queueing model', 'tandem queue'])\n", "('method used for task', 'introducing some special XXXXX related to the tollbooth XXXXX', ['performance measures', 'tandem queue'])\n", "('NONE', 'developing the XXXXX and techniques that are the cornerstones of static XXXXX', ['network flows', 'key concepts'])\n", "('NONE', 'vehicle emissions and total XXXXX are considered in the upper XXXXX', ['travel time', 'level problem'])\n", "('NONE', 'XXXXX provision does have an impact on the number of XXXXX', ['traffic information', 'optimal designs'])\n", "('NONE', 'promoters and ribosome XXXXX categorized as low , medium and XXXXX', ['binding sites', 'high efficiency'])\n", "('method used for task', 'price-lead time-inventory trade-off with XXXXX and XXXXX', ['product variety', 'uncertain demand'])\n", "('NONE', 'study of the XXXXX of the XXXXX and the feasible region', ['geometric properties', 'level sets'])\n", "('method used for task', 'approaches with XXXXX and related XXXXX and costs are rare', ['multiple products', 'setup times'])\n", "('NONE', 'XXXXX , XXXXX', ['branch-and-price algorithm', 'column generation'])\n", "('method used for task', 'the expected cost of XXXXX is computed using a XXXXX', ['stochastic model', 'system operation'])\n", "('NONE', 'it is shown that XXXXX tracking information can yield significant XXXXX', ['real time', 'cost savings'])\n", "('NONE', 'we consider currency risk management in a XXXXX of a multinational firm with XXXXX and risk sharing contracts', ['supply chain', 'risk transfer'])\n", "('NONE', 'we consider currency risk management in a XXXXX of a multinational firm with risk transfer and XXXXX', ['supply chain', 'risk sharing contracts'])\n", "('method used for task', 'we consider currency risk management in a supply chain of a multinational firm with XXXXX and XXXXX', ['risk transfer', 'risk sharing contracts'])\n", "('NONE', 'we study and compare the impacts of XXXXX and risk sharing contracts on utilities of XXXXX members', ['supply chain', 'risk transfer'])\n", "('NONE', 'we study and compare the impacts of risk transfer and XXXXX on utilities of XXXXX members', ['supply chain', 'risk sharing contracts'])\n", "('method used for task', 'we study and compare the impacts of XXXXX and XXXXX on utilities of supply chain members', ['risk transfer', 'risk sharing contracts'])\n", "('NONE', 'the newly proposed models give reliable evaluation of XXXXX in XXXXX', ['system performance', 'heavy traffic'])\n", "('method used for task', 'a XXXXX for the XXXXX problem using conditional value at risk', ['decision support system', 'vehicle replacement'])\n", "('NONE', 'identification of the XXXXX in XXXXX', ['risk drivers', 'vehicle replacement'])\n", "('NONE', 'a faster minimum XXXXX ( msi ) XXXXX is presented', ['size instance', 'identification algorithm'])\n", "('NONE', 'we compute XXXXX using a semidefinite rather than a XXXXX', ['lower bounds', 'linear relaxation'])\n", "('NONE', 'it is the first paper that tries to regard XXXXX in reverse XXXXX', ['supply chain', 'risk measure'])\n", "('NONE', 'proposed heuristic score based subset XXXXX reduces the XXXXX into polynomial growth', ['search space', 'generation process'])\n", "('NONE', 'presence of in-city XXXXX has a positive impact on XXXXX', ['spare parts', 'maintenance financial performance'])\n", "('method used for task', 'developed XXXXX to calculate throughput and other XXXXX', ['markov chain model', 'performance indicators'])\n", "('NONE', 'used XXXXX to quantify impact of XXXXX on line performance', ['markov model', 'system parameters'])\n", "('NONE', 'used XXXXX to quantify impact of system parameters on XXXXX', ['markov model', 'line performance'])\n", "('NONE', 'used markov model to quantify impact of XXXXX on XXXXX', ['system parameters', 'line performance'])\n", "('method used for task', 'XXXXX is employed to examine the effects of XXXXX modification on the prioritization solution', ['sensitivity analysis', 'parameter values'])\n", "('method used for task', 'we formulate an adapted XXXXX model to select the XXXXX', ['optimal strategy', 'mean-variance portfolio'])\n", "('NONE', 'characterizing the XXXXX utilizing XXXXX', ['sensitivity analysis', 'anchor points'])\n", "('method used for task', 'a XXXXX and a XXXXX for the operational problem are developed', ['dynamic model', 'linear approximation'])\n", "('method used for task', 'a XXXXX discusses performance with respect to XXXXX', ['computational study', 'problem characteristics'])\n", "('NONE', 'computational results show very XXXXX of the proposed XXXXX', ['high efficiency', 'planning framework'])\n", "('NONE', 'the XXXXX increases not only sales but XXXXX and default risk', ['trade credit', 'opportunity cost'])\n", "('method used for task', 'the XXXXX increases not only sales but opportunity cost and XXXXX', ['trade credit', 'default risk'])\n", "('method used for task', 'the trade credit increases not only sales but XXXXX and XXXXX', ['opportunity cost', 'default risk'])\n", "('method used for task', 'both optimal XXXXX and optimal XXXXX not only exist but also are unique', ['trade credit', 'cycle time'])\n", "('NONE', 'we study XXXXX in serial multi-tier XXXXX', ['vertical integration', 'supply chains'])\n", "('NONE', 'multiobjective binary XXXXX under XXXXX is captured', ['linear programming', 'interval uncertainty'])\n", "('method used for task', 'incorporating XXXXX and co2 emission costs into an abc XXXXX', ['capacity expansion', 'decision model'])\n", "('method used for task', 'incorporating XXXXX and XXXXX costs into an abc decision model', ['capacity expansion', 'co2 emission'])\n", "('NONE', 'incorporating capacity expansion and XXXXX costs into an abc XXXXX', ['decision model', 'co2 emission'])\n", "('NONE', 'proposing a XXXXX in selecting green XXXXX', ['mathematical programming approach', 'building projects'])\n", "('NONE', 'management must incorporate XXXXX cost into overall green XXXXX costs', ['co2 emission', 'building project'])\n", "('NONE', 'we investigate the value of accounting for XXXXX in XXXXX', ['inventory control', 'demand seasonality'])\n", "('NONE', 'we compute the XXXXX of policies which ignore part or all XXXXX', ['optimality gaps', 'demand seasonality'])\n", "('method used for task', 'considering systemicity of risks ensures more XXXXX with respect to XXXXX', ['robust decision making', 'risk analysis'])\n", "('NONE', 'building risk maps using XXXXX approaches improves XXXXX', ['problem structuring', 'risk analysis'])\n", "('method used for task', 'new sampling algorithm giving XXXXX to XXXXX is proposed', ['fast convergence', 'uniform distribution'])\n", "('method used for task', 'a hybrid heuristic algorithm based on XXXXX and XXXXX are developed', ['simulated annealing', 'binary search'])\n", "('NONE', 'XXXXX of XXXXX ( mmds ) have three levels of representation : elements , loops and the whole model', ['mental models', 'dynamic systems'])\n", "('method used for task', 'an XXXXX and a XXXXX are developed', ['branch-and-cut algorithm', 'integer programming formulation'])\n", "('NONE', 'XXXXX from 318 XXXXX in the u.s', ['survey data', 'manufacturing firms'])\n", "('NONE', 'we develop a 4-approximation algorithm for incremental XXXXX with XXXXX', ['network design', 'shortest paths'])\n", "('method used for task', 'liner fleet XXXXX solved without requiring complete XXXXX', ['probability distributions', 'deployment problem'])\n", "('NONE', 'we have studied the mobile facility routing and XXXXX with XXXXX', ['scheduling problem', 'stochastic demand'])\n", "('method used for task', 'the reliability bounds provide an XXXXX for very XXXXX', ['approximate solution', 'large networks'])\n", "('method used for task', 'XXXXX of the digraph is very effective and greatly reduces XXXXX', ['parallel processing', 'computational time'])\n", "('NONE', 'we apply our model to XXXXX of XXXXX', ['breast cancer', 'radiation therapy treatment planning'])\n", "('NONE', 'the XXXXX improves all best XXXXX known in literature', ['lower bounds', 'lower bound methodology'])\n", "('NONE', 'a better heuristic XXXXX in finding minimal XXXXX is presented', ['solution technique', 'cost bounds'])\n", "('NONE', 'our algorithm , bbh , can solve real-world XXXXX with short XXXXX', ['computation time', 'size instances'])\n", "('NONE', 'bbh outperforms the current state of the art XXXXX for most of the XXXXX', ['solution methods', 'problem instances'])\n", "('NONE', 'the XXXXX of XXXXX is extended to the case of hrs model', ['malmquist index', 'productivity change'])\n", "('NONE', 'preference model is a set of XXXXX compatible with XXXXX', ['value functions', 'preference information'])\n", "('NONE', 'XXXXX is a set of XXXXX compatible with preference information', ['value functions', 'preference model'])\n", "('NONE', 'XXXXX is a set of value functions compatible with XXXXX', ['preference information', 'preference model'])\n", "('NONE', 'we find improved best XXXXX for 18 out of the 37 XXXXX', ['lower bounds', 'benchmark instances'])\n", "('method used for task', 'the technique sequentially minimizes risk in terms of XXXXX and XXXXX', ['waiting time', 'service time'])\n", "('method used for task', 'the technique applies to any convex XXXXX and XXXXX distribution', ['loss function', 'service time'])\n", "('NONE', 'the novel XXXXX concept of line-integrated supermarkets in order to supply XXXXX is introduced', ['assembly lines', 'part logistics'])\n", "('method used for task', 'we design a XXXXX for ship stowage XXXXX', ['classification scheme', 'planning problems'])\n", "('NONE', 'a XXXXX of XXXXX is established for symmetric unimodal demand distributions', ['lower bound', 'cost deviation'])\n", "('method used for task', 'the XXXXX is more sensitive than the XXXXX to sub-optimal ordering decisions', ['newsvendor model', 'eoq model'])\n", "('method used for task', 'a new distributed XXXXX is designed for approximating the XXXXX of bi-objective optimization problems ( dlbs )', ['heuristic approach', 'pareto set'])\n", "('method used for task', 'we develop mixed XXXXX and XXXXX for the proposed models', ['exact algorithms', 'integer programs'])\n", "('NONE', 'an estimation method is proposed for XXXXX in choice based XXXXX', ['value functions', 'conjoint analysis'])\n", "('method used for task', 'an XXXXX is proposed for XXXXX in choice based conjoint analysis', ['value functions', 'estimation method'])\n", "('method used for task', 'an XXXXX is proposed for value functions in choice based XXXXX', ['conjoint analysis', 'estimation method'])\n", "('NONE', 'we consider the role of XXXXX in XXXXX in decentralized assembly networks', ['knowledge sharing', 'cost reduction'])\n", "('NONE', 'application to truncation and XXXXX in XXXXX', ['inventory models', 'supply chain management'])\n", "('NONE', 'we develop a new XXXXX , hesitant XXXXX', ['optimization method', 'fuzzy programming method'])\n", "('method used for task', 'we compare 12 heuristics and 9 XXXXX for the simple XXXXX type 1', ['lower bounds', 'assembly line balancing problem'])\n", "('method used for task', 'the success of heuristics and XXXXX strongly depends on the XXXXX', ['lower bounds', 'problem characteristics'])\n", "('method used for task', 'we employ XXXXX and XXXXX using preference for resources', ['genetic algorithm', 'combinatorial auction'])\n", "('NONE', 'we analyze how the XXXXX affects the strategies in a two-tiers XXXXX', ['spot market', 'supply chain'])\n", "('NONE', 'we study how XXXXX affect the behavior and dynamic of the XXXXX members', ['risk factors', 'supply chain'])\n", "('NONE', 'we consider a XXXXX with one global constraint on XXXXX', ['carbon emissions', 'lot-sizing problem'])\n", "('method used for task', 'both optimal XXXXX and XXXXX not only exist but also are unique', ['trade credit', 'cycle time'])\n", "('NONE', 'XXXXX of XXXXX with perturbations on weights', ['robustness analysis', 'pareto optimal solutions'])\n", "('method used for task', 'new XXXXX ( svr ) techniques are applied to XXXXX of corporate bonds', ['support vector regression', 'recovery rates'])\n", "('NONE', 'transformation of XXXXX does not improve the XXXXX', ['prediction accuracy', 'recovery rates'])\n", "('method used for task', 'a heuristic XXXXX is implemented and compared to the XXXXX', ['greedy algorithm', 'exact method'])\n", "('NONE', 'XXXXX are obtained by solving single-objective XXXXX using cplex', ['pareto solutions', 'milp models'])\n", "('method used for task', 'a XXXXX is concerned with both efficient investment and XXXXX', ['supply chain', 'production decisions'])\n", "('NONE', 'the dynamic XXXXX utilizes multiple XXXXX including fim™ instrument , bbs and moca©/mmse', ['dea model', 'assessment criteria'])\n", "('method used for task', 'we study the XXXXX for two demand classes with heterogeneous XXXXX', ['order acceptance', 'lead times'])\n", "('method used for task', 'we study the XXXXX for two XXXXX with heterogeneous lead times', ['order acceptance', 'demand classes'])\n", "('NONE', 'we study the order acceptance for two XXXXX with heterogeneous XXXXX', ['lead times', 'demand classes'])\n", "('NONE', 'XXXXX , XXXXX , reversed XXXXX and stochastic ordering comparisons are carried out', ['likelihood ratio', 'failure rate'])\n", "('NONE', 'XXXXX , XXXXX , reversed XXXXX and stochastic ordering comparisons are carried out', ['likelihood ratio', 'failure rate'])\n", "('method used for task', 'we find a XXXXX where decentralization is better than XXXXX', ['time interval', 'vertical integration'])\n", "('NONE', 'deterministic formulation where XXXXX follow a XXXXX', ['uncertain parameters', 'normal distribution'])\n", "('method used for task', 'XXXXX and an efficient XXXXX', ['branch-and-cut algorithm', 'heuristic algorithm'])\n", "('NONE', 'a XXXXX should match with manufacturers’ XXXXX ( mc )', ['consignment contract', 'sales promotion'])\n", "('NONE', 'a wholesale XXXXX should match with retailers’ XXXXX ( rp )', ['price contract', 'sales promotion'])\n", "('method used for task', 'we combine the advantages of XXXXX and XXXXX', ['evolutionary algorithms', 'statistical model'])\n", "('method used for task', 'new access charges XXXXX on the XXXXX', ['model based', 'system approach'])\n", "('method used for task', 'it is an XXXXX that does not guarantee to catch the XXXXX', ['approximation method', 'global optimum'])\n", "('method used for task', 'we examine the impact of consumer environmental awareness ( cea ) on XXXXX and XXXXX', ['channel coordination', 'order quantities'])\n", "('NONE', 'a XXXXX can effectively coordinate a decentralized XXXXX', ['return policy', 'supply chain'])\n", "('NONE', 'we study the inconsistency of XXXXX in XXXXX', ['pairwise comparisons', 'group decision making'])\n", "('NONE', 'it benefits of the XXXXX , that can be solved with modest XXXXX', ['linear relaxation', 'computational time'])\n", "('method used for task', 'proof of existence of XXXXX for stochastic linear XXXXX', ['optimal policies', 'control problems'])\n", "('method used for task', 'the method adopts the concept of lca and integrates fuzzy XXXXX and XXXXX', ['fuzzy topsis', 'extent analysis'])\n", "('NONE', 'this study considers homebuyers’ XXXXX with s-shaped XXXXX', ['risk attitude', 'utility functions'])\n", "('NONE', 'we propose a cyclical square-root continuous-time model for the XXXXX of XXXXX', ['term structure', 'interest rates'])\n", "('method used for task', 'we use XXXXX and XXXXX for the pricing subproblem', ['benders’ decomposition', 'constraint programming'])\n", "('NONE', 'the role of XXXXX on XXXXX and productivity is assessed', ['social capital', 'production efficiency'])\n", "('NONE', 'introduce a method to measure XXXXX in XXXXX', ['operational risk', 'emerging markets'])\n", "('method used for task', 'XXXXX is higher in XXXXX than developed ones', ['operational risk', 'emerging markets'])\n", "('method used for task', 'XXXXX more dominant influence on XXXXX than sector', ['operational risk', 'market development'])\n", "('NONE', 'assessment of sensor reading weights based on XXXXX using the XXXXX', ['multiple criteria', 'ahp method'])\n", "('method used for task', 'XXXXX for XXXXX and spot prices including their interrelations', ['simulation model', 'wind power'])\n", "('method used for task', 'XXXXX for wind power and XXXXX including their interrelations', ['simulation model', 'spot prices'])\n", "('method used for task', 'simulation model for XXXXX and XXXXX including their interrelations', ['wind power', 'spot prices'])\n", "('NONE', 'the XXXXX of XXXXX is strongly influenced by the dependence structure', ['market value', 'wind turbines'])\n", "('NONE', 'the XXXXX of wind turbines is strongly influenced by the XXXXX', ['market value', 'dependence structure'])\n", "('method used for task', 'the market value of XXXXX is strongly influenced by the XXXXX', ['wind turbines', 'dependence structure'])\n", "('NONE', 'provides suggested routes for XXXXX that can combine key themes in analytics and XXXXX', ['future research', 'operational research'])\n", "('NONE', 'estimating XXXXX from inspections performed at irregular XXXXX', ['dirichlet distribution', 'time intervals'])\n", "('method used for task', 'the close relationship between XXXXX and XXXXX is studied', ['aggregation functions', 'spread measures'])\n", "('NONE', 'among spread measures we find many well-known data analysis tools like the XXXXX , XXXXX , mad , range , and iqr', ['standard deviation', 'sample variance'])\n", "('NONE', 'non-symmetric XXXXX may be used e.g . in XXXXX', ['group decision making', 'spread measures'])\n", "('NONE', 'we consider the impact of various aspects which may affect the estimation of probability of XXXXX in the context of XXXXX , such as uncertainty level , alternative preference scales and different weight estimation methods . we also consider the case where the comparisons are carried out in a fuzzy manner . it is revealed that the results hold regardless the aforementioned aspects considered', ['rank reversal', 'pairwise comparisons'])\n", "('NONE', 'we consider the impact of various aspects which may affect the estimation of probability of XXXXX in the context of pairwise comparisons , such as XXXXX , alternative preference scales and different weight estimation methods . we also consider the case where the comparisons are carried out in a fuzzy manner . it is revealed that the results hold regardless the aforementioned aspects considered', ['rank reversal', 'uncertainty level'])\n", "('NONE', 'we consider the impact of various aspects which may affect the estimation of probability of rank reversal in the context of XXXXX , such as XXXXX , alternative preference scales and different weight estimation methods . we also consider the case where the comparisons are carried out in a fuzzy manner . it is revealed that the results hold regardless the aforementioned aspects considered', ['pairwise comparisons', 'uncertainty level'])\n", "('method used for task', 'we use convex XXXXX to combine scheduling and XXXXX', ['non-linear programming', 'capacity planning'])\n", "('method used for task', 'we provide an XXXXX that captures the key trade-offs among time-to-market , XXXXX , pricing and production decisions', ['optimization model', 'sales channel'])\n", "('method used for task', 'we provide an XXXXX that captures the key trade-offs among time-to-market , sales channel , pricing and XXXXX', ['optimization model', 'production decisions'])\n", "('method used for task', 'we provide an optimization model that captures the key trade-offs among time-to-market , XXXXX , pricing and XXXXX', ['sales channel', 'production decisions'])\n", "('NONE', 'we characterize XXXXX , XXXXX and production policies', ['optimal pricing', 'market timing'])\n", "('NONE', 'we study the role of the secondary XXXXX on XXXXX , pricing and inventory decisions', ['sales channel', 'market timing'])\n", "('method used for task', 'we study the role of the secondary XXXXX on market timing , pricing and XXXXX', ['sales channel', 'inventory decisions'])\n", "('method used for task', 'we study the role of the secondary sales channel on XXXXX , pricing and XXXXX', ['market timing', 'inventory decisions'])\n", "('method used for task', 'we provide a method to characterize XXXXX for XXXXX', ['optimal policies', 'optimal stopping problems'])\n", "('NONE', \"we model a product 's degradation path by a wiener stochastic process and investigate the XXXXX of a step-stress accelerated degradation test ( ssadt ) for obtaining precise estimates of XXXXX\", ['optimum design', 'model parameters'])\n", "('NONE', 'changes in XXXXX of discrete XXXXX over time investigated', ['survival models', 'parameter estimates'])\n", "('method used for task', 'an improved h-mk-svm is developed to integrate the external , tag and keyword , individual behavioral and engagement behavioral data for XXXXX from multiple correlated attributes and XXXXX', ['feature selection', 'ensemble learning'])\n", "('NONE', 'consider uncertainty and XXXXX in XXXXX first time', ['multiple objectives', 'traveling salesman problem'])\n", "('NONE', 'we analyze a situation in which several firms deal with XXXXX concerning the same type of product and having XXXXX warehouses', ['limited capacity', 'inventory problems'])\n", "('NONE', 'we model a period-review XXXXX with XXXXX and fixed order costs', ['lost sales', 'inventory system'])\n", "('NONE', 'XXXXX considers the impact of XXXXX on financial efficiency', ['dynamic model', 'environmental variables'])\n", "('NONE', 'by focusing on linear and affine XXXXX we justify the use of new XXXXX', ['random variables', 'uncertainty sets'])\n", "('method used for task', 'the zone based XXXXX is modeled as a XXXXX', ['vehicle routing problem', 'multi-objective problem'])\n", "('method used for task', 'XXXXX asymmetry is confined to developed markets during XXXXX downtowns only', ['stock market', 'return reversal'])\n", "('method used for task', 'significantly improve best-known XXXXX for XXXXX', ['solution values', 'benchmark problem instances'])\n", "('method used for task', 'working environment affects operators’ XXXXX and the XXXXX', ['system performance', 'health condition'])\n", "('method used for task', 'work-related ill health ( wih ) XXXXX are considered to model XXXXX of the operators', ['risk factors', 'health states'])\n", "('method used for task', 'we develop efficient XXXXX based on XXXXX , vlsn search , and a hybrid algorithm that integrates the two', ['heuristic algorithms', 'tabu search'])\n", "('method used for task', 'we develop efficient XXXXX based on tabu search , vlsn search , and a XXXXX that integrates the two', ['heuristic algorithms', 'hybrid algorithm'])\n", "('method used for task', 'we develop efficient heuristic algorithms based on XXXXX , vlsn search , and a XXXXX that integrates the two', ['tabu search', 'hybrid algorithm'])\n", "('NONE', 'the XXXXX establishes that effective integration of simple XXXXX with vlsn search results in superior outcomes', ['computational study', 'tabu search'])\n", "('method used for task', 'we propose a dynamic pricing model with XXXXX with different behaviors applied to the XXXXX', ['multiple products', 'airline industry'])\n", "('NONE', 'we propose a XXXXX with XXXXX with different behaviors applied to the airline industry', ['multiple products', 'dynamic pricing model'])\n", "('method used for task', 'we propose a XXXXX with multiple products with different behaviors applied to the XXXXX', ['airline industry', 'dynamic pricing model'])\n", "('NONE', 'XXXXX eats green product’s sales more under certain XXXXX', ['limited capacity', 'market conditions'])\n", "('method used for task', 'a modified silver–meal heuristic combined with a XXXXX is proposed to overcome XXXXX', ['safety stock', 'demand variability'])\n", "('NONE', 'we study XXXXX policies under which consumers’ valuation depends on both the refund amount and the length of the XXXXX', ['consumer returns', 'return deadline'])\n", "('method used for task', 'we put forward a new differentiated buy-back contract contingent on the XXXXX to coordinate the XXXXX', ['supply chain', 'return deadline'])\n", "('NONE', 'new type of indirect XXXXX in form of assignment-based XXXXX of alternatives', ['preference information', 'pairwise comparisons'])\n", "('NONE', 'applications on XXXXX as well as on two XXXXX show that the proposed approaches outperform traditional techniques for conjoint analysis', ['experimental data', 'empirical studies'])\n", "('method used for task', 'applications on XXXXX as well as on two empirical studies show that the proposed approaches outperform traditional techniques for XXXXX', ['experimental data', 'conjoint analysis'])\n", "('method used for task', 'applications on experimental data as well as on two XXXXX show that the proposed approaches outperform traditional techniques for XXXXX', ['empirical studies', 'conjoint analysis'])\n", "('NONE', 'validating the proposed model and XXXXX via a real XXXXX', ['case study', 'solution technique'])\n", "('NONE', 'we design traditional XXXXX using different XXXXX', ['classification methods', 'failure models'])\n", "('method used for task', 'we propose a new cost-efficiency measure based on the notions of ray XXXXX and optimal XXXXX', ['average cost', 'scale size'])\n", "('method used for task', \"the new XXXXX ( ace ) extends banker and thrall 's ( 1992 ) tsre measure to XXXXX\", ['cost analysis', 'efficiency measure'])\n", "('method used for task', 'we show new lower and XXXXX on the XXXXX of spt', ['upper bounds', 'worst-case ratio'])\n", "('NONE', 'we structure the field of XXXXX in the XXXXX', ['automotive industry', 'part logistics'])\n", "('method used for task', 'XXXXX based on XXXXX is a latest problem in d–s theory', ['parameter estimation', 'belief structure'])\n", "('method used for task', 'XXXXX based on interval-valued XXXXX is first studied', ['parameter estimation', 'belief structure'])\n", "('method used for task', 'an incremental XXXXX is studied , where the objective is to maximize the cumulative s-t-flow over a XXXXX', ['network design problem', 'time horizon'])\n", "('NONE', 'the XXXXX as well as the heuristics are compared in XXXXX on randomly generated instances', ['computational experiments', 'mip formulations'])\n", "('NONE', 'we develop an minlp model which determines the XXXXX of a mixed ac–dc building electrical XXXXX', ['optimal design', 'distribution system'])\n", "('NONE', 'the XXXXX with simultaneous pickup and delivery with XXXXX is presented and the literature is summarized', ['vehicle routing problem', 'time limit'])\n", "('NONE', 'a number of well-known XXXXX with XXXXX and no XXXXX are solved and the solutions are compared', ['benchmark problems', 'time limit'])\n", "('NONE', 'a number of well-known XXXXX with XXXXX and no XXXXX are solved and the solutions are compared', ['benchmark problems', 'time limit'])\n", "('method used for task', 'a perturbation based variable neighborhood search heuristic proposed new best and XXXXX for XXXXX', ['robust solutions', 'benchmark problem instances'])\n", "('NONE', 'employing the XXXXX in XXXXX without taking into account that the distribution is mixed normal leads to a loss of 2–3 percent per year', ['portfolio selection', 'mean-variance rule'])\n", "('NONE', 'the XXXXX is determined from the variability of each indicator projected onto XXXXX', ['principal components', 'preference structure'])\n", "('method used for task', 'nonlinear XXXXX are suitable for agricultural XXXXX', ['policy analysis', 'mean-variance models'])\n", "('method used for task', 'a two-stage XXXXX with an order-up-to XXXXX is considered', ['supply chain', 'inventory policy'])\n", "('NONE', 'two new procedures using a multi-attribute XXXXX with incomplete XXXXX on weights related to performance measures', ['utility function', 'preference information'])\n", "('method used for task', 'two new procedures using a multi-attribute XXXXX with incomplete preference information on weights related to XXXXX', ['utility function', 'performance measures'])\n", "('method used for task', 'two new procedures using a multi-attribute utility function with incomplete XXXXX on weights related to XXXXX', ['preference information', 'performance measures'])\n", "('NONE', 'a XXXXX of risk propagation in a XXXXX is developed', ['supply network', 'bayesian network model'])\n", "('NONE', 'XXXXX demonstrates how the measures are used in a XXXXX setting', ['simulation study', 'supply network'])\n", "('NONE', 'we propose an XXXXX of a two-echelon recycled pulp XXXXX', ['agent-based model', 'supply chain'])\n", "('method used for task', 'XXXXX for u-shaped XXXXX are investigated', ['optimization models', 'assembly lines'])\n", "('NONE', 'we test the benefits of XXXXX , optimizing the periodic review interval , versus XXXXX , changing to a continuous review policy', ['parameter optimization', 'policy improvement'])\n", "('NONE', 'XXXXX outperforms XXXXX in most instances tested', ['parameter optimization', 'policy improvement'])\n", "('NONE', 'XXXXX for solutions of multi objective XXXXX are proposed', ['ranking methods', 'optimization problems'])\n", "('method used for task', 'we propose an XXXXX for the minimum common string XXXXX', ['ilp model', 'partition problem'])\n", "('method used for task', 'semi-open jackson XXXXX for XXXXX', ['network model', 'problem formulation'])\n", "('method used for task', 'XXXXX to find the XXXXX', ['iterative algorithm', 'optimal threshold'])\n", "('method used for task', 'the modelization provides very good XXXXX especially for rank 2 XXXXX matrix', ['lp relaxation', 'separation time'])\n", "('method used for task', 'XXXXX leads to a XXXXX based on constraint generation', ['best approximation', 'heuristic method'])\n", "('method used for task', 'XXXXX leads to a heuristic method based on XXXXX', ['best approximation', 'constraint generation'])\n", "('method used for task', 'best approximation leads to a XXXXX based on XXXXX', ['heuristic method', 'constraint generation'])\n", "('NONE', 'in context , XXXXX of several problems of XXXXX is analyzed', ['computational complexity', 'simple games'])\n", "('NONE', 'several XXXXX with XXXXX are presented sequentially over time', ['multiple attributes', 'decision alternatives'])\n", "('NONE', 'we extend the use of the combined stochastic dea–bayesian model to examine the XXXXX of XXXXX generated through the metafrontier', ['statistical properties', 'efficiency scores'])\n", "('method used for task', 'first a XXXXX on the XXXXX is shown ; then an exponential-time online optimal algorithm for the problem is presented ; and lastly a polynomial-time online asymptotically optimal algorithm is presented', ['lower bound', 'competitive ratio'])\n", "('method used for task', 'first a XXXXX on the competitive ratio is shown ; then an exponential-time online XXXXX for the problem is presented ; and lastly a polynomial-time online asymptotically XXXXX is presented', ['lower bound', 'optimal algorithm'])\n", "('method used for task', 'first a lower bound on the XXXXX is shown ; then an exponential-time online XXXXX for the problem is presented ; and lastly a polynomial-time online asymptotically XXXXX is presented', ['competitive ratio', 'optimal algorithm'])\n", "('NONE', 'experimental results show that the polynomial-time XXXXX takes only a fraction of the XXXXX of the offline optimal algorithm , yet produces solutions of competitive quality to the offline optimal solutions', ['online algorithm', 'running time'])\n", "('method used for task', 'we combine XXXXX and XXXXX approaches', ['stochastic programming', 'stochastic optimal control'])\n", "('NONE', 'we generate XXXXX with XXXXX and events of death', ['scenario trees', 'asset returns'])\n", "('NONE', 'the choice of the XXXXX at the customization point affects the total XXXXX', ['service level', 'operating cost'])\n", "('NONE', 'the choice of the XXXXX at the XXXXX affects the total operating cost', ['service level', 'customization point'])\n", "('NONE', 'the choice of the service level at the XXXXX affects the total XXXXX', ['operating cost', 'customization point'])\n", "('NONE', 'a non-constant XXXXX impacts the XXXXX', ['service level', 'customization point'])\n", "('NONE', 'focusing on determining the XXXXX with minimum total XXXXX in a dsm', ['activity sequence', 'feedback time'])\n", "('NONE', 'we solve two routing and XXXXX with split pickup and XXXXX', ['scheduling problems', 'split delivery'])\n", "('NONE', 'the discrete XXXXX provides quicker results but less XXXXX', ['split model', 'solution quality'])\n", "('method used for task', 'we provide an XXXXX to minimize the total delay cost , yielding XXXXX in short time', ['exact algorithm', 'exact solutions'])\n", "('method used for task', 'novel adaptation of efficient XXXXX to XXXXX', ['global optimization', 'dynamic environments'])\n", "('method used for task', 'the sensitivity to different XXXXX and XXXXX is studied', ['model parameters', 'cost parameters'])\n", "('NONE', 'XXXXX on response and recovery planning phases of XXXXX', ['literature survey', 'disaster management'])\n", "('NONE', 'for the XXXXX with a buffer , an XXXXX is developed for solving it', ['scheduling problem', 'optimal algorithm'])\n", "('method used for task', 'we derive XXXXX to solve the problems in XXXXX', ['sufficient conditions', 'polynomial time'])\n", "('method used for task', 'two metaheuristics are proposed based on XXXXX and XXXXX', ['neighborhood search', 'tree search'])\n", "('method used for task', 'we consider identical XXXXX to minimize XXXXX', ['parallel machine scheduling', 'total tardiness'])\n", "('method used for task', 'XXXXX and a XXXXX solve small problems', ['mathematical programming', 'branch-and-bound algorithm'])\n", "('method used for task', 'a XXXXX and a XXXXX are developed for larger problems', ['tabu search', 'hybrid genetic algorithm'])\n", "('NONE', 'XXXXX in XXXXX can lead to lower order rates', ['information sharing', 'supply chains'])\n", "('method used for task', 'a capital XXXXX for XXXXX has been suggested', ['coherent risk measures', 'allocation scheme'])\n", "('method used for task', 'focus on lagrangian decomposition , XXXXX , and XXXXX', ['column generation', 'benders’ decomposition'])\n", "('NONE', 'XXXXX increases not only sales but also XXXXX and default risk', ['trade credit', 'opportunity cost'])\n", "('method used for task', 'XXXXX increases not only sales but also opportunity cost and XXXXX', ['trade credit', 'default risk'])\n", "('method used for task', 'trade credit increases not only sales but also XXXXX and XXXXX', ['opportunity cost', 'default risk'])\n", "('method used for task', 'we use XXXXX to XXXXX in prices and network capacity', ['stochastic programming', 'model uncertainty'])\n", "('method used for task', 'we use XXXXX to model uncertainty in prices and XXXXX', ['stochastic programming', 'network capacity'])\n", "('NONE', 'we use stochastic programming to XXXXX in prices and XXXXX', ['model uncertainty', 'network capacity'])\n", "('NONE', 'XXXXX on the loss from delaying exercise of XXXXX are found', ['upper bounds', 'american options'])\n", "('NONE', 'we develop a novel XXXXX of XXXXX', ['probabilistic model', 'audit quality'])\n", "('NONE', 'we find that firms often adopt a XXXXX a XXXXX after its introduction', ['new technology', 'time lag'])\n", "('NONE', 'we study stochastic XXXXX , a class of stochastic XXXXX', ['cooperative games', 'linear programming games'])\n", "('NONE', 'for some certainty equivalents , a XXXXX can be computed in XXXXX', ['polynomial time', 'core allocation'])\n", "('NONE', 'approach a new pickup and XXXXX with lifo and XXXXX', ['time constraints', 'delivery problem'])\n", "('NONE', 'evaluate the performance of the proposed XXXXX through a comprehensive XXXXX', ['computational study', 'solution methods'])\n", "('method used for task', 'we propose a new XXXXX model for XXXXX', ['option pricing', 'random volatility'])\n", "('method used for task', 'an XXXXX is developed to solve the polynomial XXXXX', ['efficient algorithm', 'optimization problem'])\n", "('NONE', 'output XXXXX causes ex post XXXXX', ['price uncertainty', 'profit loss'])\n", "('NONE', 'XXXXX and technical inefficiency are the main sources of XXXXX', ['risk preference', 'profit loss'])\n", "('method used for task', 'we also consider a XXXXX and provide XXXXX on the number of pmus', ['theoretical model', 'lower bounds'])\n", "('NONE', 'the algorithm can achieve the optimum in the XXXXX of the XXXXX', ['theoretical model', 'power systems'])\n", "('NONE', 'we developed an XXXXX that solves the XXXXX', ['exact algorithm', 'feature selection problem'])\n", "('method used for task', 'considerable improvements in XXXXX and XXXXX over earlier methods', ['running time', 'solution quality'])\n", "('NONE', 'g-anp is specially useful for XXXXX with XXXXX', ['big data', 'decision making problems'])\n", "('NONE', 'it is applicable to the production of XXXXX in a XXXXX', ['multiple products', 'job shop'])\n", "('NONE', 'introduction of XXXXX extending the notion of XXXXX', ['aggregation functions', 'fusion functions'])\n", "('NONE', 'we provide a time window XXXXX in a vrp setting with XXXXX', ['demand uncertainty', 'assignment model'])\n", "('NONE', 'XXXXX show that using multiple XXXXX is better than one', ['numerical experiments', 'demand scenarios'])\n", "('method used for task', 'they are applicable to any number of XXXXX and any XXXXX', ['demand classes', 'demand distribution'])\n", "('method used for task', 'we consider sustainability issues on the joint XXXXX and XXXXX', ['trade credit', 'inventory decisions'])\n", "('NONE', 'XXXXX can improve consumer confidence and XXXXX', ['traceability systems', 'supply chain profit'])\n", "('NONE', 'we analyze the impact of XXXXX on XXXXX', ['failure interaction', 'warranty cost'])\n", "('method used for task', 'we carry out a real XXXXX to illustrate the XXXXX proposed', ['case study', 'multicriteria model'])\n", "('method used for task', 'we define an XXXXX for the bi-objective bi-dimensional XXXXX', ['knapsack problem', 'upper bound set'])\n", "('method used for task', 'XXXXX is relevant to tactical multi-project XXXXX', ['capacity planning', 'resource loading'])\n", "('NONE', 'to study a book XXXXX through XXXXX', ['supply chain', 'mathematical modeling'])\n", "('method used for task', 'the impact of XXXXX and different XXXXX is investigated', ['load distribution', 'objective functions'])\n", "('method used for task', 'the scope of or practice has been extended via XXXXX ( soft or ) and XXXXX', ['problem structuring methods', 'business analytics'])\n", "('NONE', 'to address how to improve the XXXXX in a two-level decentralized XXXXX with random yield and uncertain demand', ['service level', 'supply chain'])\n", "('NONE', 'to address how to improve the XXXXX in a two-level decentralized supply chain with XXXXX and uncertain demand', ['service level', 'random yield'])\n", "('NONE', 'to address how to improve the XXXXX in a two-level decentralized supply chain with random yield and XXXXX', ['service level', 'uncertain demand'])\n", "('NONE', 'to address how to improve the service level in a two-level decentralized XXXXX with XXXXX and uncertain demand', ['supply chain', 'random yield'])\n", "('NONE', 'to address how to improve the service level in a two-level decentralized XXXXX with random yield and XXXXX', ['supply chain', 'uncertain demand'])\n", "('method used for task', 'to address how to improve the service level in a two-level decentralized supply chain with XXXXX and XXXXX', ['random yield', 'uncertain demand'])\n", "('NONE', 'we provide the XXXXX of XXXXX for all the problems', ['lower bounds', 'competitive ratio'])\n", "('method used for task', 'development of various XXXXX and XXXXX', ['lower bounds', 'dominance rules'])\n", "('method used for task', 'a XXXXX is formulated to minimize the total XXXXX', ['operating cost', 'dynamic programming model'])\n", "('method used for task', 'XXXXX and a XXXXX are developed', ['branch-and-bound algorithm', 'milp formulations'])\n", "('method used for task', 'XXXXX is currently the best XXXXX', ['branch-and-bound algorithm', 'exact method'])\n", "('method used for task', 'we develop a pseudo-polynomial XXXXX for concave XXXXX', ['time algorithm', 'compression cost function'])\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "('method used for task', 'we develop a XXXXX for convex XXXXX', ['polynomial time algorithm', 'compression cost function'])\n", "('method used for task', 'we develop a new XXXXX for minmax regret XXXXX', ['lower bound', 'combinatorial optimization problems'])\n", "('method used for task', 'our solution reduces two costs linked to this problem : XXXXX and handling XXXXX', ['fuel consumption', 'operations costs'])\n", "('NONE', 'the XXXXX can be found by a XXXXX', ['dynamic programming algorithm', 'optimal routing strategy'])\n", "('NONE', 'the XXXXX has a specific XXXXX', ['optimal strategy', 'threshold structure'])\n", "('NONE', 'XXXXX and the value of the marginal product of inputs are measured in one step using various XXXXX', ['technical efficiency', 'model specifications'])\n", "('NONE', 'we use a bootstrap dea technique to estimate the mean and 95 percent XXXXX of XXXXX and shadow prices', ['confidence intervals', 'technical efficiency'])\n", "('NONE', 'we use a bootstrap dea technique to estimate the mean and 95 percent XXXXX of technical efficiency and XXXXX', ['confidence intervals', 'shadow prices'])\n", "('method used for task', 'we use a bootstrap dea technique to estimate the mean and 95 percent confidence intervals of XXXXX and XXXXX', ['technical efficiency', 'shadow prices'])\n", "('NONE', 'we show how to approximate the XXXXX via XXXXX', ['efficient frontier', 'integer programming'])\n", "('method used for task', 'XXXXX to optimise biomass-based XXXXX', ['mathematical model', 'supply chains'])\n", "('NONE', 'an XXXXX with XXXXX where positive lead times are allowed', ['inventory model', 'random yield'])\n", "('NONE', 'an XXXXX with random yield where positive XXXXX are allowed', ['inventory model', 'lead times'])\n", "('NONE', 'an inventory model with XXXXX where positive XXXXX are allowed', ['random yield', 'lead times'])\n", "('NONE', 'we introduce an XXXXX of optimal value and XXXXX', ['upper bound', 'heuristic algorithms'])\n", "('method used for task', 'we state new production axioms that account for XXXXX and formally derive XXXXX from them', ['ratio data', 'dea models'])\n", "('method used for task', 'establish that expected loss in general stochastic XXXXX is non-increasing as XXXXX measured using entropy is reduced', ['demand uncertainty', 'inventory problems'])\n", "('NONE', 'classified by , e.g. , XXXXX , input data , objective , XXXXX', ['model formulation', 'solution method'])\n", "('method used for task', 'we present mean-variance models for hybrid XXXXX and translate them into XXXXX', ['portfolio optimization', 'convex quadratic programming'])\n", "('method used for task', 'we consider the XXXXX and give the XXXXX in the case with no more than two new securities', ['analytical solutions', 'solution procedures'])\n", "('method used for task', 'empirical estimates indicate difference in the parameter coefficients of gamma stochastic XXXXX , and heterogeneity function variables between the pooled and the swamy–arora XXXXX', ['production function', 'panel estimator'])\n", "('NONE', 'consider the XXXXX with XXXXX , multi-factor volatility and jumps', ['mean reversion', 'asset dynamics'])\n", "('NONE', 'we derive XXXXX to various types of discrete XXXXX', ['closed-form solutions', 'variance swaps'])\n", "('NONE', 'the XXXXX with a small XXXXX rate prefers the pooling strategy', ['service provider', 'demand loss'])\n", "('NONE', 'the XXXXX with a small demand loss rate prefers the XXXXX', ['service provider', 'pooling strategy'])\n", "('NONE', 'the service provider with a small XXXXX rate prefers the XXXXX', ['demand loss', 'pooling strategy'])\n", "('NONE', 'we describe the search region in XXXXX by local XXXXX', ['multi-objective optimization', 'upper bounds'])\n", "('NONE', 'we describe the XXXXX in XXXXX by local upper bounds', ['multi-objective optimization', 'search region'])\n", "('NONE', 'we describe the XXXXX in multi-objective optimization by local XXXXX', ['upper bounds', 'search region'])\n", "('NONE', 'we reveal XXXXX of the optimal instalment payment , which is different from that under XXXXX', ['structural properties', 'moral hazard'])\n", "('NONE', 'we consider a XXXXX with a piecewise linear XXXXX', ['transportation problem', 'cost structure'])\n", "('method used for task', 'we propose a XXXXX for stowage XXXXX of a container ship', ['heuristic algorithm', 'planning problem'])\n", "('method used for task', 'we give a framework for XXXXX and dynamic XXXXX', ['risk measures', 'stochastic optimization problems'])\n", "('method used for task', 'the XXXXX ( mopath ) is embedded in XXXXX', ['fuzzy model', 'optimization framework'])\n", "('method used for task', 'approximation based on XXXXX provides good solutions and XXXXX', ['network flows', 'lower bounds'])\n", "('NONE', 'XXXXX , dynamic , XXXXX , heterogeneous fleet constraints are handled', ['time windows', 'split delivery'])\n", "('method used for task', 'a XXXXX is also developed for larger instances which produces good XXXXX', ['heuristic procedure', 'approximate solutions'])\n", "('NONE', 'we propose a XXXXX that takes into account both cancellations and XXXXX behaviour', ['revenue management model', 'customer choice'])\n", "('method used for task', 'we develop an XXXXX for scheduling chp units with XXXXX', ['efficient algorithm', 'electric vehicles'])\n", "('method used for task', 'XXXXX for XXXXX with redials and reconnects ;', ['queueing model', 'call centers'])\n", "('NONE', 'we aim to maximise shareholder wealth while mitigating XXXXX of XXXXX in manufacturing', ['environmental impact', 'carbon emission'])\n", "('method used for task', 'studies in joint configuration of XXXXX and XXXXX are reviewed', ['supply chains', 'product families'])\n", "('method used for task', 'XXXXX for product families and XXXXX are formulated', ['mathematical models', 'supply chains'])\n", "('method used for task', 'XXXXX for XXXXX and supply chains are formulated', ['mathematical models', 'product families'])\n", "('method used for task', 'mathematical models for XXXXX and XXXXX are formulated', ['supply chains', 'product families'])\n", "('NONE', 'the socp approximations are used to obtain XXXXX at the top levels of a branch and XXXXX', ['lower bounds', 'bound algorithm'])\n", "('method used for task', 'we propose a three-stage XXXXX e-nautilus for computationally expensive XXXXX', ['interactive method', 'multiobjective optimization problems'])\n", "('NONE', 'a set of pre-calculated XXXXX enables a solution process without XXXXX', ['pareto optimal solutions', 'waiting times'])\n", "('NONE', 'a set of pre-calculated XXXXX enables a XXXXX without waiting times', ['pareto optimal solutions', 'solution process'])\n", "('NONE', 'a set of pre-calculated pareto optimal solutions enables a XXXXX without XXXXX', ['waiting times', 'solution process'])\n", "('method used for task', 'no new XXXXX is solved when the XXXXX interacts with the method', ['optimization problem', 'decision maker'])\n", "('method used for task', 'first result to comprehensively determine the XXXXX under endogenous and exogenous XXXXX', ['optimal policy', 'demand uncertainty'])\n", "('NONE', 'a XXXXX of published XXXXX for managing supply chain risks', ['systematic review', 'analytical models'])\n", "('NONE', 'a XXXXX of published analytical models for managing XXXXX', ['systematic review', 'supply chain risks'])\n", "('method used for task', 'a systematic review of published XXXXX for managing XXXXX', ['analytical models', 'supply chain risks'])\n", "('NONE', 'eight XXXXX for formal modeling of XXXXX', ['research streams', 'supply chain risks'])\n", "('method used for task', 'we propose a strong XXXXX ( mip ) for the stand XXXXX', ['mixed integer programming', 'allocation problem'])\n", "('NONE', 'it is shown that XXXXX can solve efficiently the XXXXX', ['greedy algorithms', 'view selection problem'])\n", "('NONE', 'the XXXXX perturbs a solution using XXXXX', ['simulated annealing', 'solution approach'])\n", "('method used for task', 'we present a XXXXX to solve resource-constrained XXXXX with flexible project structure', ['genetic algorithm', 'scheduling problems'])\n", "('method used for task', 'we present a XXXXX to solve resource-constrained scheduling problems with flexible XXXXX', ['genetic algorithm', 'project structure'])\n", "('NONE', 'we present a genetic algorithm to solve resource-constrained XXXXX with flexible XXXXX', ['scheduling problems', 'project structure'])\n", "('NONE', \"XXXXX of the shape of the firm 's XXXXX is presented\", ['empirical assessment', 'cost function'])\n", "('NONE', 'XXXXX are obtained under exponential XXXXX', ['closed-form solutions', 'utility function'])\n", "('NONE', 'we propose an approach for achieving XXXXX in XXXXX', ['customer satisfaction', 'manufacturing firms'])\n", "('NONE', 'a XXXXX was conducted in 24 XXXXX from 3 countries', ['questionnaire survey', 'manufacturing firms'])\n", "('NONE', 'the model is fitted to univariate inflow XXXXX by XXXXX', ['time series', 'quantile regression'])\n", "('method used for task', 'the XXXXX is to transfer all data in given time at the XXXXX', ['scheduling problem', 'minimum cost'])\n", "('NONE', 'advance notice , XXXXX , XXXXX and correlation are considered', ['lead times', 'random yield'])\n", "('NONE', 'longer remanufacturing XXXXX could reduce XXXXX', ['lead times', 'inventory variance'])\n", "('NONE', 'optimizing the removal of a XXXXX has low impact on XXXXX', ['base station', 'survival probability'])\n", "('NONE', 'we study the effect of information updating in a XXXXX with XXXXX', ['supply chain', 'spot market'])\n", "('method used for task', 'the new XXXXX also updates the belief on the XXXXX , and vice versa', ['demand information', 'spot price'])\n", "('method used for task', 'giving a XXXXX for obtaining the XXXXX in fdh models', ['polynomial-time algorithm', 'anchor points'])\n", "('NONE', 'find the role of XXXXX and time horizon in the probability of adoption of XXXXX', ['initial conditions', 'renewable energy'])\n", "('method used for task', 'find the role of XXXXX and XXXXX in the probability of adoption of renewable energy', ['initial conditions', 'time horizon'])\n", "('NONE', 'find the role of initial conditions and XXXXX in the probability of adoption of XXXXX', ['renewable energy', 'time horizon'])\n", "('NONE', 'trade resistance and XXXXX modeled as an undesirable XXXXX', ['production process', 'trade barriers'])\n", "('method used for task', 'a new XXXXX is also suggested for two XXXXX', ['calibration procedure', 'factor models'])\n", "('method used for task', 'we propose an XXXXX to optimize the lru XXXXX', ['milp model', 'definition decision problem'])\n", "('NONE', 'presenting XXXXX based on proven properties of the XXXXX', ['optimal algorithms', 'cost function'])\n", "('NONE', 'we tested the algorithm on XXXXX from the XXXXX', ['uci repository', 'benchmark instances'])\n", "('NONE', 'we compare XXXXX , XXXXX and deterministic approaches', ['robust optimization', 'stochastic programming'])\n", "('NONE', 'the XXXXX scales well with the size of the XXXXX', ['power system', 'robust optimization approach'])\n", "('NONE', 'models for the XXXXX of a district XXXXX were developed', ['optimal design', 'cooling system'])\n", "('NONE', 'XXXXX with XXXXX dependent cost function', ['parallel machine scheduling', 'completion time'])\n", "('NONE', 'XXXXX with completion time dependent XXXXX', ['parallel machine scheduling', 'cost function'])\n", "('NONE', 'parallel machine scheduling with XXXXX dependent XXXXX', ['completion time', 'cost function'])\n", "('method used for task', 'XXXXX solved via novel XXXXX', ['decomposition method', 'mixed integer programming model'])\n", "('method used for task', 'we develop a behavioral XXXXX for revenue sharing contracts incorporating XXXXX', ['decision model', 'reference points'])\n", "('method used for task', 'we conduct three XXXXX to test our XXXXX', ['laboratory experiments', 'behavioral model'])\n", "('method used for task', 'we model the problem with XXXXX and XXXXX', ['mixed integer programming', 'constraint programming'])\n", "('method used for task', 'bfp has lower XXXXX and better XXXXX than lpe', ['computational complexity', 'worst-case performance'])\n", "('method used for task', 'the multi-period XXXXX is extended by considering XXXXX goals', ['vehicle routing problem', 'service consistency'])\n", "('method used for task', 'a 70 percent better XXXXX is achieved by increasing XXXXX by not more than 3.84 percent , on average', ['travel time', 'arrival time consistency'])\n", "('NONE', 'formulated a XXXXX of two XXXXX simultaneously', ['decision problem', 'decision variables'])\n", "('method used for task', 'the impact on distance , XXXXX , and XXXXX are investigated', ['fuel consumption', 'labor costs'])\n", "('method used for task', 'introduction of a multi-attribute XXXXX approach for XXXXX', ['panel data', 'peer assessment'])\n", "('NONE', 'the risk-neutral XXXXX exhibit non-zero XXXXX of variance risk', ['diffusion limits', 'market prices'])\n", "('NONE', 'the risk-neutral XXXXX exhibit non-zero market prices of XXXXX', ['diffusion limits', 'variance risk'])\n", "('NONE', 'the risk-neutral diffusion limits exhibit non-zero XXXXX of XXXXX', ['market prices', 'variance risk'])\n", "('NONE', 'the XXXXX covers 80 XXXXX , 518 threats , and 1244 safeguards', ['knowledge base', 'system components'])\n", "('NONE', 'we present the XXXXX with XXXXX', ['vehicle routing problem', 'release dates'])\n", "('NONE', 'we study XXXXX in an unknown , changing , XXXXX', ['dynamic pricing', 'market environment'])\n", "('NONE', 'XXXXX with quasi-arithmetic means and XXXXX', ['aggregation operators', 'linguistic information'])\n", "('NONE', 'we develop XXXXX ( with XXXXX o ( n2log ( n ) ) for both cases', ['exact algorithms', 'time complexity'])\n", "('method used for task', 'the XXXXX indicator is decomposed into XXXXX ;', ['industry inefficiency', 'sources components'])\n", "('NONE', 'we present a new XXXXX with family setups and XXXXX', ['resource constraints', 'single-machine scheduling problem'])\n", "('method used for task', 'we present a hybrid procedure combining two XXXXX for solving the XXXXX', ['exact methods', 'pricing problem'])\n", "('method used for task', 'our method outperforms an existing branch-and-cut both in XXXXX and XXXXX', ['cpu time', 'solution quality'])\n", "('method used for task', 'group model building is a XXXXX to XXXXX', ['participatory approach', 'system dynamics'])\n", "('method used for task', 'XXXXX is a XXXXX to system dynamics', ['participatory approach', 'group model building'])\n", "('method used for task', 'XXXXX is a participatory approach to XXXXX', ['system dynamics', 'group model building'])\n", "('NONE', 'use the XXXXX with shortages allowed under limited XXXXX', ['eoq model', 'storage capacity'])\n", "('NONE', 'the method used to calculate physical resource tightness is improved by setting a XXXXX of XXXXX', ['critical value', 'resource availability'])\n", "('method used for task', 'the method used to calculate physical XXXXX is improved by setting a XXXXX of resource availability', ['critical value', 'resource tightness'])\n", "('NONE', 'the method used to calculate physical XXXXX is improved by setting a critical value of XXXXX', ['resource availability', 'resource tightness'])\n", "('method used for task', 'an innovative application of XXXXX to analyse hiv XXXXX', ['operations research', 'gene sequences'])\n", "('method used for task', 'XXXXX for XXXXX with 35 , 973 , 840 states were solved', ['optimal control problems', 'production systems'])\n", "('method used for task', 'first bi/multi-objective model for the XXXXX routing and XXXXX', ['home care', 'scheduling problem'])\n", "('method used for task', 'a XXXXX based on multi-directional XXXXX', ['metaheuristic algorithm', 'local search'])\n", "('NONE', 'XXXXX on new XXXXX based on real-life data', ['numerical experiments', 'benchmark instances'])\n", "('method used for task', 'we present the first-ever XXXXX for the consistent XXXXX', ['exact method', 'traveling salesman problem'])\n", "('method used for task', 'we propose novel XXXXX and associated XXXXX', ['valid inequalities', 'separation techniques'])\n", "('NONE', 'XXXXX behaviours are modelled using XXXXX', ['markov models', 'treatment adherence'])\n", "('method used for task', 'we show minimal XXXXX and XXXXX do not lead to minimum costs', ['safety stock', 'inventory variance'])\n", "('method used for task', 'we describe properties of the XXXXX function and compute valid lower and XXXXX', ['upper bounds', 'service time'])\n", "('NONE', 'we examine single-stage dea models with XXXXX that include a very small XXXXX on weights', ['weight restrictions', 'lower bound'])\n", "('NONE', 'we examine single-stage XXXXX with XXXXX that include a very small lower bound on weights', ['weight restrictions', 'dea models'])\n", "('NONE', 'we examine single-stage XXXXX with weight restrictions that include a very small XXXXX on weights', ['lower bound', 'dea models'])\n", "('method used for task', 'green XXXXX programs have many dimensions and XXXXX to consider', ['supplier development', 'big data sets'])\n", "('NONE', 'directions for XXXXX on green XXXXX investment are presented', ['future research', 'supplier development'])\n", "('method used for task', 'minimax regret and XXXXX are used to address XXXXX', ['stochastic programming', 'demand uncertainty'])\n", "('NONE', 'a XXXXX justifies the XXXXX provided by proposed model', ['computational study', 'robust solution'])\n", "('NONE', 'bgeva on woe method for XXXXX showed the best XXXXX', ['missing data', 'predictive accuracy'])\n", "('method used for task', 'a XXXXX for the salesmen is defined by means of a cooperative game theory XXXXX , i.e . the ( least ) core', ['cost allocation', 'solution concept'])\n", "('NONE', 'we model and develop algorithms to reduce XXXXX over XXXXX', ['large-scale networks', 'network size'])\n", "('NONE', 'XXXXX generates more eco-friendly XXXXX under specific conditions', ['supply chain coordination', 'social welfare'])\n", "('method used for task', 'we use XXXXX and XXXXX to solve the problem', ['mixed integer programming', 'benders decomposition'])\n", "('NONE', 'we study the XXXXX of our XXXXX in different setups', ['computation time', 'decomposition algorithm'])\n", "('NONE', 'decision makers’ XXXXX were taken into account in the XXXXX', ['decision process', 'risk preferences'])\n", "('NONE', 'accelerated XXXXX with XXXXX and pareto-optimal cuts', ['benders decomposition', 'valid inequalities'])\n", "('method used for task', 'XXXXX for process mean , pricing , production and XXXXX', ['integrated framework', 'market segmentation'])\n", "('NONE', 'integrating XXXXX with XXXXX in a single framework', ['loss function', 'posterior probability'])\n", "('method used for task', 'using optimized XXXXX and XXXXX', ['monte carlo simulation', 'hybrid genetic algorithm'])\n", "('NONE', 'in a great britain XXXXX , we find that a XXXXX increases generation adequacy', ['case study', 'capacity market'])\n", "('NONE', 'we develop XXXXX of XXXXX to implement ‘stsd’', ['linear systems', 'optimality conditions'])\n", "('method used for task', 'organizations engage in routine XXXXX and XXXXX', ['decision making', 'problem solving'])\n", "('method used for task', 'routine XXXXX and XXXXX have different implications for or', ['decision making', 'problem solving'])\n", "('method used for task', 'our XXXXX based on the branch and price principle solve the multi-trip XXXXX with time windows', ['exact methods', 'vehicle routing problem'])\n", "('method used for task', 'our XXXXX based on the branch and price principle solve the multi-trip vehicle routing problem with XXXXX', ['exact methods', 'time windows'])\n", "('NONE', 'our exact methods based on the branch and price principle solve the multi-trip XXXXX with XXXXX', ['vehicle routing problem', 'time windows'])\n", "('NONE', 'we identify a XXXXX using the XXXXX given by cplex', ['pareto front', 'solution pool'])\n", "('NONE', 'XXXXX with stochastic demands and XXXXX is investigated', ['vehicle routing problem', 'time windows'])\n", "('NONE', 'we show how hard XXXXX can be modeled as XXXXX', ['control problems', 'integer programs'])\n", "('NONE', 'we provide a general framework for searching XXXXX in XXXXX', ['parameter space', 'monte carlo simulation'])\n", "('method used for task', 'XXXXX and lower and XXXXX are added to speed up the algorithm', ['upper bounds', 'dominance rules'])\n", "('NONE', 'the XXXXX outperforms a commercial XXXXX on small instances', ['exact method', 'ilp solver'])\n", "('NONE', 'study dynamic XXXXX with multiple XXXXX in continuous-time', ['portfolio selection', 'risk measures'])\n", "('NONE', 'method is tested using real XXXXX from a london XXXXX', ['road network', 'traffic data'])\n", "('NONE', 'including boosted trees , XXXXX , penalised linear/semi-parametric XXXXX', ['random forests', 'logistic regression'])\n", "('NONE', 'XXXXX suggest boosted XXXXX outperform penalised logistic regression', ['statistical tests', 'regression trees'])\n", "('NONE', 'XXXXX suggest boosted regression trees outperform penalised XXXXX', ['statistical tests', 'logistic regression'])\n", "('NONE', 'statistical tests suggest boosted XXXXX outperform penalised XXXXX', ['regression trees', 'logistic regression'])\n", "('NONE', 'we propose a XXXXX and a comprehensive set of new XXXXX', ['tabu search', 'benchmark instances'])\n", "('NONE', 'an automatic XXXXX as the engine of the dynamic adaptive XXXXX', ['feedback mechanism', 'consensus model'])\n", "('NONE', 'the XXXXX may be replaced by XXXXX', ['standard deviation', 'coherent risk measures'])\n", "('NONE', 'we propose the variable intermediate measures sbm model ( vsbm ) to evaluate the XXXXX of XXXXX', ['system efficiency', 'two-stage processes'])\n", "('method used for task', 'XXXXX to approximate european XXXXX are proposed', ['closed-form formulas', 'option prices'])\n", "('method used for task', 'a XXXXX is applied to XXXXX and u.s. government bond yields', ['calibration procedure', 'option prices'])\n", "('method used for task', 'we use a XXXXX to update the XXXXX under repeated measurements', ['bayesian approach', 'system state'])\n", "('NONE', 'an embedded XXXXX which deals with XXXXX', ['multiple objectives', 'local search procedure'])\n", "('NONE', 'the XXXXX of XXXXX is improved', ['upper bound', 'competitive ratio'])\n", "('method used for task', 'linear-fractional programming and primal-dual XXXXX are used to find the XXXXX of a central inequality', ['upper bound', 'analysis methods'])\n", "('NONE', 'smart and efficient XXXXX composed by auxiliary XXXXX', ['neighborhood structures', 'data structures'])\n", "('NONE', 'ten XXXXX consulting XXXXX', ['system dynamics', 'case studies'])\n", "('method used for task', 'an inverse XXXXX finds a coherent risk measure for a given XXXXX', ['optimal portfolio', 'portfolio problem'])\n", "('method used for task', 'an inverse portfolio problem finds a XXXXX measure for a given XXXXX', ['optimal portfolio', 'coherent risk'])\n", "('NONE', 'an inverse XXXXX finds a XXXXX measure for a given optimal portfolio', ['portfolio problem', 'coherent risk'])\n", "('NONE', 'necessary and XXXXX for the existence of such a XXXXX are obtained', ['sufficient conditions', 'risk measure'])\n", "('method used for task', 'XXXXX ( fs ) is modelled as a ( mixed ) XXXXX', ['feature selection', 'integer optimization problem'])\n", "('NONE', 'the accumulation of biases can cause XXXXX in XXXXX', ['path dependence', 'decision analysis'])\n", "('NONE', 'the framework is XXXXX of XXXXX', ['expected utility', 'profit maximization'])\n", "('NONE', 'autocorrelation , XXXXX , and long order cycles can degrade XXXXX', ['lead time', 'inventory performance'])\n", "('method used for task', 'two new XXXXX are employed to reduce the XXXXX', ['search space', 'dominance rules'])\n", "('method used for task', 'we develop an improved XXXXX based on a XXXXX', ['approximation algorithm', 'minimum spanning tree'])\n", "('method used for task', 'an XXXXX based on a XXXXX is presented for the problem', ['exact algorithm', 'decomposition scheme'])\n", "('method used for task', 'a XXXXX based on a XXXXX is developed', ['heuristic algorithm', 'mixed integer programming'])\n", "('NONE', 'XXXXX disclosed the superiority of XXXXX', ['experimental analysis', 'hybrid algorithms'])\n", "('method used for task', 'cohort XXXXX is applied for solving three XXXXX', ['combinatorial problems', 'intelligence method'])\n", "('NONE', 'we derive a new improved XXXXX of 2.618 for the XXXXX for the online strip packing', ['lower bound', 'competitive ratio'])\n", "('method used for task', 'we derive a new improved XXXXX of 2.618 for the competitive ratio for the online XXXXX', ['lower bound', 'strip packing'])\n", "('method used for task', 'we derive a new improved lower bound of 2.618 for the XXXXX for the online XXXXX', ['competitive ratio', 'strip packing'])\n", "('method used for task', 'we develop a XXXXX based on stochastic dual XXXXX', ['dynamic programming', 'solution approach'])\n", "('method used for task', 'simple XXXXX are summarized by the XXXXX', ['decision rules', 'optimal strategy'])\n", "('method used for task', 'improvements in XXXXX and XXXXX over earlier approaches', ['running time', 'solution quality'])\n", "('NONE', 'decide the rescue team assignment scheme by solving the second XXXXX using the proposed XXXXX', ['heuristic algorithm', 'stage model'])\n", "('NONE', 'network design models augmented with XXXXX provide XXXXX', ['valid inequalities', 'lower bounds'])\n", "('method used for task', 'XXXXX are reported on large XXXXX', ['computational experiments', 'problem instances'])\n", "('NONE', 'the XXXXX involves XXXXX in terms of pairwise comparisons', ['interactive procedure', 'preference information'])\n", "('NONE', 'the XXXXX involves preference information in terms of XXXXX', ['interactive procedure', 'pairwise comparisons'])\n", "('NONE', 'the interactive procedure involves XXXXX in terms of XXXXX', ['preference information', 'pairwise comparisons'])\n", "('NONE', 'generate XXXXX given region-specific characteristics via XXXXX', ['robust solutions', 'decision analysis'])\n", "('method used for task', 'a functional ito’s XXXXX is adopted to evaluate XXXXX', ['calculus approach', 'convex risk measures'])\n", "('method used for task', 'we develop a theory of XXXXX due to XXXXX', ['risk aversion', 'resource dependency'])\n", "('NONE', 'we show that XXXXX significantly impacts optimal XXXXX', ['investment decisions', 'resource dependency'])\n", "('NONE', 'uncertainty and XXXXX are XXXXX for resource dependency', ['risk aversion', 'sufficient conditions'])\n", "('method used for task', 'uncertainty and XXXXX are sufficient conditions for XXXXX', ['risk aversion', 'resource dependency'])\n", "('method used for task', 'uncertainty and risk aversion are XXXXX for XXXXX', ['sufficient conditions', 'resource dependency'])\n", "('NONE', 'we formulate and solve the pickup and XXXXX with XXXXX and multiple stacks', ['time windows', 'delivery problem'])\n", "('NONE', 'two branch-price-and-cut algorithms are implemented to solve the pickup and XXXXX with XXXXX and multiple stacks', ['time windows', 'delivery problem'])\n", "('method used for task', 'new XXXXX and XXXXX are developed', ['lower bounds', 'dominance rules'])\n", "('method used for task', 'a XXXXX is suggested to transform XXXXX into a linear one', ['heuristic algorithm', 'nonlinear model'])\n", "('NONE', 'we model a XXXXX with XXXXX from competitors', ['competitive environment', 'incomplete data'])\n", "('method used for task', 'the assignment of decision objects to decision classes is based on the use of the XXXXX and XXXXX', ['majority principle', 'veto effect'])\n", "('method used for task', 'the XXXXX and XXXXX notions are implemented through the concepts of concordance and discordance', ['majority principle', 'veto effect'])\n", "('NONE', 'coelli et al.’s ( 2007 ) environmental XXXXX does not reward XXXXX', ['efficiency measure', 'pollution control'])\n", "('NONE', 'a new environmental XXXXX that rewards XXXXX efforts is proposed', ['efficiency measure', 'pollution control'])\n", "('NONE', 'a XXXXX identifies cases solvable in XXXXX', ['complexity analysis', 'polynomial time'])\n", "('NONE', 'item rearrangements along XXXXX bring significant routing XXXXX', ['vehicle routes', 'cost savings'])\n", "('method used for task', 'we examine a parallel machine scheduling problem with XXXXX and XXXXX', ['setup times', 'time windows'])\n", "('method used for task', 'XXXXX ( ip ) and XXXXX ( cp ) models are proposed', ['integer programming', 'constraint programming'])\n", "('NONE', 'the XXXXX of the XXXXX is on https : //github.com/ctu-iig/nrrpgpu', ['source code', 'parallel algorithm'])\n", "('NONE', 'we present a XXXXX based XXXXX', ['benders decomposition', 'solution method'])\n", "('NONE', 'XXXXX when dual XXXXX outperforms single bundle price', ['numerical analysis', 'pricing strategy'])\n", "('NONE', 'using a XXXXX , our model can be solved with a standard XXXXX', ['decomposition approach', 'ip solver'])\n", "('method used for task', 'this XXXXX shock models based on a new XXXXX of shocks', ['counting process', 'paper studies'])\n", "('method used for task', 'XXXXX and consistency of the mle is established for XXXXX', ['asymptotic normality', 'mle metamodels'])\n", "('NONE', 'we study XXXXX and advertising policies in a XXXXX for a monopolist firm', ['dynamic pricing', 'sales model'])\n", "('method used for task', 'we consider the XXXXX and XXXXX in two machine flow shops', ['order acceptance', 'scheduling problem'])\n", "('method used for task', 'we use XXXXX to model spatial/functional connectivity in XXXXX', ['integer programming', 'site selection'])\n", "('method used for task', 'we present a XXXXX for the size-robust XXXXX', ['knapsack problem', 'column generation algorithm'])\n", "('method used for task', 'we present a XXXXX for the demand robust XXXXX', ['shortest path problem', 'column generation algorithm'])\n", "('method used for task', 'we show that XXXXX decomposes uniquely into technical efficiency change and XXXXX', ['technical change', 'productivity change'])\n", "('NONE', 'we derive an aggregate XXXXX from individual XXXXX', ['productivity index', 'productivity indices'])\n", "('NONE', 'XXXXX based XXXXX on the quality of service are devised', ['mathematical programming', 'lower bounds'])\n", "('NONE', 'we develop novel heuristics for XXXXX in XXXXX', ['channel assignment', 'wireless mesh networks'])\n", "('method used for task', 'we review the recent advances in XXXXX for XXXXX , minlp', ['global optimization', 'mixed integer nonlinear programming'])\n", "('method used for task', 'an XXXXX is applied to a XXXXX with profits', ['adaptive large neighborhood search', 'vehicle routing problem'])\n", "('method used for task', 'the XXXXX were successfully applied to a XXXXX and their worth is demonstrated', ['case study', 'expansion models'])\n", "('NONE', 'XXXXX : all phases of XXXXX', ['disaster management', 'application area'])\n", "('NONE', 'XXXXX are classified and discussed , as well as XXXXX', ['objective functions', 'solution methods'])\n", "('NONE', 'we consider a XXXXX with XXXXX on unrelated parallel machines', ['scheduling model', 'maintenance activities'])\n", "('NONE', 'analytic and XXXXX of the two-stage XXXXX are presented', ['numerical solutions', 'dynamic model'])\n", "('method used for task', 'q-g and q-cg are q-analogs of the XXXXX and XXXXX', ['steepest descent', 'conjugate gradient methods'])\n", "('NONE', 'an efficient XXXXX that considers the XXXXX and “solvency-based” measures to produce offspring', ['scatter search', 'search history'])\n", "('NONE', 'we use a XXXXX to estimate the distribution of XXXXX times between any two locations in a city', ['bayesian method', 'ambulance travel'])\n", "('method used for task', 'our XXXXX estimates outperform a recently-published method and a commercial XXXXX', ['travel time', 'software package'])\n", "('NONE', 'the XXXXX of nonlinear XXXXX is continuous with respect to changing the measure', ['value function', 'stochastic optimization problems'])\n", "('method used for task', 'algorithm to minimize XXXXX for tree-shaped XXXXX', ['investment cost', 'gas distribution networks'])\n", "('method used for task', 'we extend the XXXXX to take into account a XXXXX of the set of criteria', ['hierarchical structure', 'electre tri methods'])\n", "('method used for task', 'the XXXXX hierarchy process is applied in conjunction with the XXXXX', ['multiple criteria', 'electre tri methods'])\n", "('method used for task', 'two XXXXX based on johnsons rule and XXXXX', ['approximation algorithms', 'linear programming'])\n", "('NONE', 'XXXXX of 4 XXXXX of pmp is conducted', ['comparative study', 'exact solution methods'])\n", "('method used for task', 'develop algorithm to compute XXXXX for discrete XXXXX', ['robust solutions', 'optimisation problems'])\n", "('NONE', 'we present a cyclic production scheme for XXXXX with XXXXX', ['multiple products', 'stochastic demand'])\n", "('NONE', 'we consider XXXXX , XXXXX and storage capacity', ['sequence-dependent setup times', 'service levels'])\n", "('method used for task', 'we consider XXXXX , service levels and XXXXX', ['sequence-dependent setup times', 'storage capacity'])\n", "('method used for task', 'we consider sequence-dependent setup times , XXXXX and XXXXX', ['service levels', 'storage capacity'])\n", "('NONE', 'the XXXXX based on real world data compares six strategies that control the XXXXX', ['computational study', 'cycle length'])\n", "('NONE', 'we present a biobjective formulation for identifying XXXXX from a given XXXXX', ['robust solutions', 'pareto set'])\n", "('method used for task', 'XXXXX and a solution algorithm are developed for the case of multiobjective XXXXX', ['structural properties', 'linear programs'])\n", "('method used for task', 'XXXXX and a XXXXX are developed for the case of multiobjective linear programs', ['structural properties', 'solution algorithm'])\n", "('NONE', 'structural properties and a XXXXX are developed for the case of multiobjective XXXXX', ['linear programs', 'solution algorithm'])\n", "('method used for task', 'we derive equivalent XXXXX . thus , we compute optimal solutions and XXXXX', ['upper bounds', 'mixed integer linear programming formulations'])\n", "('NONE', 'a XXXXX illustrates XXXXX of defensive resources', ['case study', 'optimal allocation'])\n", "('method used for task', 'we take a XXXXX and develop a XXXXX', ['branch-and-price algorithm', 'column generation approach'])\n", "('NONE', 'we provide a competitive XXXXX of the online one-way trading problem with XXXXX on prices', ['limited information', 'difference analysis'])\n", "('NONE', 'the gas market is represented by a partial XXXXX including XXXXX', ['market power', 'equilibrium model'])\n", "('NONE', 'we develop a XXXXX embedding fairness concerns into the XXXXX', ['behavioral model', 'utility functions'])\n", "('method used for task', 'we propose a novel XXXXX , considering unloading activities at the terminal and XXXXX scheduling', ['integrated model', 'pipeline pumping'])\n", "('NONE', 'we propose a methodology based on XXXXX , XXXXX and a method to obtain optimal scenario trees', ['historical data', 'statistical analysis'])\n", "('method used for task', 'we propose a methodology based on XXXXX , statistical analysis and a method to obtain optimal XXXXX', ['historical data', 'scenario trees'])\n", "('method used for task', 'we propose a methodology based on historical data , XXXXX and a method to obtain optimal XXXXX', ['statistical analysis', 'scenario trees'])\n", "('method used for task', 'XXXXX for XXXXX at isolated intersections are derived', ['stochastic models', 'queue lengths'])\n", "('method used for task', 'we analyze XXXXX for XXXXX projects , accounting for market risk preferences', ['investment timing', 'risk reduction'])\n", "('NONE', 'a mechanism to discard low XXXXX before the XXXXX is proposed', ['local search', 'quality solutions'])\n", "('NONE', 'combination of binary XXXXX and minimization of XXXXX', ['decision trees', 'boolean functions'])\n", "('method used for task', 'identifies the selection of XXXXX for XXXXX as research question', ['business analytics', 'data items'])\n", "('NONE', 'we model the XXXXX problem with XXXXX for container departure', ['time windows', 'container relocation'])\n", "('NONE', 'our approach accounts for dependencies among XXXXX and uncertainty in XXXXX', ['project scores', 'portfolio constraints'])\n", "('method used for task', 'equivalence of surcharge pricing with joint economic XXXXX and XXXXX has been established', ['lot sizing', 'quantity discount'])\n", "('method used for task', 'we study the two-machine flowshop problem with sequence-independent XXXXX to minimize XXXXX', ['setup times', 'total completion time'])\n", "('NONE', 'we study the XXXXX with sequence-independent XXXXX to minimize total completion time', ['setup times', 'two-machine flowshop problem'])\n", "('method used for task', 'we study the XXXXX with sequence-independent setup times to minimize XXXXX', ['total completion time', 'two-machine flowshop problem'])\n", "('NONE', 'XXXXX assess the efficiency of XXXXX', ['numerical experiments', 'branch-and-bound algorithms'])\n", "('method used for task', 'we present a XXXXX for XXXXX', ['supplier selection', 'stochastic programming model'])\n", "('NONE', 'solving a single XXXXX in the XXXXX is more efficient', ['benders decomposition', 'master problem'])\n", "('NONE', 'we examine the effects of service and XXXXX on XXXXX', ['optimal policy', 'competition intensity'])\n", "('method used for task', 'we propose an XXXXX for an XXXXX', ['exact algorithm', 'np-hard problem'])\n", "('method used for task', 'we characterize the structure of its XXXXX and derive XXXXX', ['optimal policy', 'analytical solution'])\n", "('method used for task', 'a XXXXX for XXXXX and economic manufacturing quantity is proposed', ['condition-based maintenance', 'joint optimization model'])\n", "('NONE', 'the assumption that the XXXXX has discrete XXXXX has been relaxed', ['state space', 'degradation process'])\n", "('NONE', 'the XXXXX from the XXXXX are considerable when the costs related to unexpected failures are high', ['cost savings', 'joint optimization model'])\n", "('method used for task', 'we study XXXXX where the processing times are also XXXXX', ['scheduling problems', 'decision variables'])\n", "('method used for task', 'co-opetition leads to more profit and less XXXXX based on higher XXXXX and increased unit XXXXX', ['carbon emissions', 'product prices'])\n", "('method used for task', 'XXXXX is used to find the minimum XXXXX', ['multiobjective optimization', 'separation values'])\n", "('NONE', 'presented a risk shaping model for XXXXX under XXXXX', ['random yield', 'production planning problem'])\n", "('NONE', 'used XXXXX as the XXXXX', ['conditional value-at-risk', 'risk measure'])\n", "('NONE', 'XXXXX increases price , whereas XXXXX reduces it', ['demand uncertainty', 'cost uncertainty'])\n", "('NONE', 'this paper raises the problem of measuring XXXXX of XXXXX in dea models', ['partial derivatives', 'transformation functions'])\n", "('NONE', 'this paper raises the problem of measuring XXXXX of transformation functions in XXXXX', ['partial derivatives', 'dea models'])\n", "('NONE', 'this paper raises the problem of measuring partial derivatives of XXXXX in XXXXX', ['transformation functions', 'dea models'])\n", "('method used for task', 'we propose a new XXXXX for solving multi-objective XXXXX', ['exact algorithm', 'integer programs'])\n", "('NONE', 'analysis the impact of XXXXX on XXXXX', ['capacity planning', 'problem characteristics'])\n", "('method used for task', 'we present a XXXXX ( de ) for XXXXX', ['modified differential evolution', 'dynamic optimization'])\n", "('NONE', 'results show that the XXXXX can exhibit chaos when XXXXX are at work', ['social interactions', 'demand dynamics'])\n", "('method used for task', 'in numerical study , we show that the XXXXX is superior to the traditional XXXXX', ['empirical bayes method', 'estimation method'])\n", "('method used for task', 'a heuristic XXXXX for this new XXXXX was developed', ['optimization problem', 'solution method'])\n", "('method used for task', 'we analyze XXXXX for the erlang-a XXXXX', ['risk measures', 'queueing model'])\n", "('NONE', 'we analyze XXXXX of queues using XXXXX', ['performance measures', 'risk measures'])\n", "('method used for task', 'we examine the effect of XXXXX and XXXXX on four types of supply contracts', ['market share', 'asymmetric information'])\n", "('NONE', 'the optimal transfer payment of a buyer is influenced by XXXXX of the buyer and XXXXX', ['market share', 'supply network structure'])\n", "('method used for task', 'uncertainty in XXXXX and the underlying XXXXX is modeled', ['probability distribution', 'risk preferences'])\n", "('method used for task', 'the XXXXX is computed exactly with XXXXX in a special case', ['optimal solution', 'convex optimization'])\n", "('method used for task', 'XXXXX are conducted for the newsvendor and XXXXX to analyze the implications of our model', ['numerical experiments', 'portfolio optimization problem'])\n", "('NONE', 'we investigate the XXXXX of XXXXX and forward-start options', ['model risk', 'variance swaps'])\n", "('NONE', 'we provide evidence that the XXXXX of XXXXX is around 100 basis points', ['model risk', 'variance swaps'])\n", "('method used for task', 'we propose a XXXXX to determine open-loop XXXXX', ['nash equilibria', 'search scheme'])\n", "('NONE', 'we give a new definition of the XXXXX inspired by the weak link in XXXXX', ['system efficiency', 'supply chains'])\n", "('NONE', 'the XXXXX a three-player XXXXX with an infinite time horizon', ['differential game', 'paper studies'])\n", "('NONE', 'the paper studies a three-player XXXXX with an infinite XXXXX', ['differential game', 'time horizon'])\n", "('NONE', 'the XXXXX a three-player differential game with an infinite XXXXX', ['paper studies', 'time horizon'])\n", "('method used for task', 'we propose a new XXXXX for solving tri-objective XXXXX', ['exact algorithm', 'integer programs'])\n", "('NONE', 'several unique XXXXX of an XXXXX are proved', ['structural properties', 'optimal schedule'])\n", "('method used for task', 'a fork-join XXXXX is formulated to analyze the XXXXX', ['queueing network', 'system performance'])\n", "('NONE', 'the XXXXX has advantages in small XXXXX', ['parallel policy', 'size systems'])\n", "('method used for task', 'we propose an XXXXX with efficient heuristics to explore the XXXXX', ['adaptive large neighborhood search', 'solution space'])\n", "('NONE', 'we provide an improved wolf XXXXX ( wsa ) using a global XXXXX', ['search algorithm', 'memory structure'])\n", "('method used for task', 'XXXXX for XXXXX and generation capacity investment', ['transmission line', 'equilibrium model'])\n", "('NONE', 'we perform a XXXXX on instances adapted from vrp XXXXX', ['computational study', 'benchmark instances'])\n", "('method used for task', 'the effect of XXXXX on the maximum XXXXX is analyzed', ['stochastic demand', 'inventory distribution'])\n", "('method used for task', 'the XXXXX take past , present and future XXXXX into account', ['performance indicators', 'dea models'])\n", "('method used for task', 'we propose a XXXXX for this novel XXXXX', ['objective function', 'dynamic pricing model'])\n", "('NONE', 'we study channel XXXXX of a dual-channel XXXXX facing potential supply shortage', ['supply chain', 'priority strategy'])\n", "('NONE', 'we examine the effects of coordination/decision sequence of XXXXX on XXXXX', ['channel priority', 'priority strategy'])\n", "('method used for task', 'pricing and XXXXX strategies are robust to the time sequence of XXXXX decision', ['channel priority', 'channel strategies'])\n", "('method used for task', 'pricing and XXXXX strategies are robust to the time sequence of XXXXX decision', ['channel priority', 'channel decision'])\n", "('method used for task', 'pricing and XXXXX strategies are robust to the time sequence of XXXXX decision', ['channel strategies', 'channel priority'])\n", "('method used for task', 'pricing and XXXXX strategies are robust to the time sequence of XXXXX decision', ['channel priority', 'channel decision'])\n", "('method used for task', 'this XXXXX the optimal preventive maintenance problem based on a new XXXXX', ['counting process', 'paper studies'])\n", "('method used for task', 'we study admission and XXXXX for a make-to-order XXXXX', ['inventory control', 'production system'])\n", "('NONE', 'we investigate the XXXXX under non-preemptive XXXXX and setup cost', ['optimal policy', 'lead time'])\n", "('method used for task', 'we investigate the XXXXX under non-preemptive lead time and XXXXX', ['optimal policy', 'setup cost'])\n", "('method used for task', 'we investigate the optimal policy under non-preemptive XXXXX and XXXXX', ['lead time', 'setup cost'])\n", "('method used for task', 'a modified wordnet based XXXXX for XXXXX disambiguation', ['similarity measure', 'word sense'])\n", "('method used for task', 'two new XXXXX are proposed based on joint XXXXX', ['mutual information', 'feature selection methods'])\n", "('method used for task', 'the best XXXXX and the lowest XXXXX are achieved', ['time complexity', 'prediction performance'])\n", "('NONE', 'uncertain XXXXX can be included in the XXXXX for aerospace panels', ['boundary conditions', 'design process'])\n", "('method used for task', 'we propose a level set-based XXXXX to design acoustic metamaterials using the XXXXX', ['topology optimization', 'finite element method'])\n", "('NONE', 'we formulate the XXXXX using effective XXXXX', ['optimization problem', 'material properties'])\n", "('NONE', 'a two-step XXXXX enabled to handle the design of effective XXXXX', ['optimization process', 'bulk modulus'])\n", "('method used for task', 'algorithms for calculating the XXXXX and the internal XXXXX are presented', ['stiffness matrix', 'force vector'])\n", "('method used for task', 'we address an isogeometric XXXXX for thermal XXXXX of functionally graded material plates', ['finite element formulation', 'buckling analysis'])\n", "('NONE', 'analysis of plate and XXXXX may involve inclined principal XXXXX', ['shell structures', 'stress fields'])\n", "('method used for task', 'a new XXXXX for steel–concrete side-plated beams in XXXXX is presented', ['finite element model', 'fire conditions'])\n", "('method used for task', 'the proposed element has a simple formulation meanwhile it shows satisfactory XXXXX and possesses XXXXX', ['high accuracy', 'convergence performance'])\n", "('method used for task', 'a refined 18-dof triangular hybrid XXXXX is given out , and the element captures the scale effect and has XXXXX', ['high accuracy', 'stress element'])\n", "('method used for task', 'interpolet-based XXXXX are obtained and presented graphically . expressions for XXXXX and force vector are derived', ['shape functions', 'stiffness matrix'])\n", "('method used for task', 'interpolet-based XXXXX are obtained and presented graphically . expressions for stiffness matrix and XXXXX are derived', ['shape functions', 'force vector'])\n", "('method used for task', 'interpolet-based shape functions are obtained and presented graphically . expressions for XXXXX and XXXXX are derived', ['stiffness matrix', 'force vector'])\n", "('NONE', 'XXXXX are represented as discrete point load at the XXXXX', ['muscle forces', 'integration points'])\n", "('NONE', 'XXXXX enables the use of XXXXX , fully integrated solid tetrahedral finite elements', ['parallel processing', 'high order'])\n", "('NONE', 'XXXXX enables the use of high order , fully integrated solid tetrahedral XXXXX', ['parallel processing', 'finite elements'])\n", "('NONE', 'parallel processing enables the use of XXXXX , fully integrated solid tetrahedral XXXXX', ['high order', 'finite elements'])\n", "('NONE', 'evolutionary XXXXX are derived from XXXXX', ['experimental measurements', 'power spectra'])\n", "('NONE', 'condensation technique of degree of freedom is first proposed to improve the XXXXX of XXXXX with galerkin weak form', ['computational efficiency', 'meshfree method'])\n", "('NONE', 'adoption of modified crack closure integral ( mcci ) in XXXXX for evaluation of stress XXXXX ( sifs )', ['meshfree methods', 'intensity factors'])\n", "('method used for task', 'XXXXX based on XXXXX is consistent with the field results', ['fe modal analysis', 'vehicle waveforms'])\n", "('method used for task', 'XXXXX based on XXXXX can be used to guide bridge inspectors', ['fe modal analysis', 'vehicle waveforms'])\n", "('NONE', 'XXXXX is simplified through a number of observed XXXXX', ['model construction', 'statistical properties'])\n", "('NONE', 'calculate the exact XXXXX between different stress modes using our new XXXXX', ['inner product', 'similarity degrees'])\n", "('NONE', 'calculate the exact similarity degrees between different XXXXX using our new XXXXX', ['inner product', 'stress modes'])\n", "('NONE', 'calculate the exact XXXXX between different XXXXX using our new inner product', ['similarity degrees', 'stress modes'])\n", "('NONE', 'derive the basic XXXXX from XXXXX and broken them into a set of sub-modes', ['displacement field', 'stress modes'])\n", "('method used for task', 'XXXXX is a method called numerical microscope in XXXXX and numerical analysis', ['wavelet analysis', 'signal processing'])\n", "('method used for task', 'XXXXX is a method called numerical microscope in signal processing and XXXXX', ['wavelet analysis', 'numerical analysis'])\n", "('method used for task', 'wavelet analysis is a method called numerical microscope in XXXXX and XXXXX', ['signal processing', 'numerical analysis'])\n", "('NONE', 'an efficient XXXXX of 3d semi-rigid XXXXX is proposed', ['steel frames', 'analysis procedure'])\n", "('method used for task', 'a XXXXX to address uncertainty effects in the XXXXX is presented', ['hybrid method', 'impact force identification'])\n", "('method used for task', 'a XXXXX is conducted to analyze the effects of material uncertainty on XXXXX of composite stiffened panels', ['parametric study', 'dynamic responses'])\n", "('NONE', 'influence of the XXXXX kinetics was taken into account by implementing in the fem model XXXXX data', ['phase transformation', 'phase transformation kinetics'])\n", "('NONE', 'XXXXX procedure at XXXXX is based on diffuse approximation', ['field transfer', 'integration points'])\n", "('NONE', 'it has a correct rank , is free from XXXXX , with no signs of XXXXX', ['spurious modes', 'shear locking'])\n", "('NONE', 'thermal behavior coupled with XXXXX in comprehensive XXXXX', ['structural analysis', 'fe model'])\n", "('method used for task', 'similar to the end shortening strain commonly used for compression , the skewed XXXXX is uniquely proposed for the first time to control the in-plane shear action in the finite XXXXX', ['angle strain', 'strip method'])\n", "('method used for task', 'when the ratio of skewed XXXXX vs. end shortening strain is large enough , the average tensile XXXXX along the longitudinal direction are produced due to the large out-of-plane deflection', ['angle strain', 'section forces'])\n", "('method used for task', 'a new XXXXX for XXXXX shear deformable composite beams is proposed', ['finite element', 'higher order'])\n", "('method used for task', 'a new XXXXX for higher order shear deformable XXXXX is proposed', ['finite element', 'composite beams'])\n", "('NONE', 'a new finite element for XXXXX shear deformable XXXXX is proposed', ['higher order', 'composite beams'])\n", "('method used for task', 'XXXXX and XXXXX can be used to calculate fracture toughness', ['exact method', 'finite element'])\n", "('method used for task', 'XXXXX and finite element can be used to calculate XXXXX', ['exact method', 'fracture toughness'])\n", "('method used for task', 'exact method and XXXXX can be used to calculate XXXXX', ['finite element', 'fracture toughness'])\n", "('method used for task', 'the framework is based on coupled XXXXX and XXXXX', ['genetic algorithm', 'finite element analysis'])\n", "('method used for task', 'XXXXX and XXXXX , insensitive mesh distortions , free volume locking , etc . are found for the cq4', ['high accuracy', 'convergence rate'])\n", "('NONE', 'XXXXX show the good performance of the proposed XXXXX', ['numerical examples', 'multiscale method'])\n", "('method used for task', 'XXXXX based on the laplace transform method are provided for the XXXXX', ['analytical solutions', 'case studies'])\n", "('method used for task', 'the XXXXX is used to construct a material XXXXX', ['density approximation', 'shepard function'])\n", "('NONE', 'XXXXX formulated as independent from angle and number of XXXXX', ['stiffness matrix', 'fourier series'])\n", "('method used for task', 'a new hybrid inactive/quiet XXXXX is proposed for modeling XXXXX', ['additive manufacturing', 'element method'])\n", "('NONE', 'XXXXX of hydrostatic XXXXX', ['performance characteristics', 'thrust bearing'])\n", "('NONE', 'XXXXX of parallel XXXXX independent of element order', ['computational time', 'matrix factorization'])\n", "('method used for task', 'fundamental XXXXX cases are investigated for XXXXX welded aluminum joints', ['impact analysis', 'friction stir'])\n", "('NONE', 'we investigate a deterministic and statistical XXXXX in XXXXX', ['size effect', 'reinforced concrete beams'])\n", "('method used for task', 'we propose a XXXXX based on the XXXXX', ['finite element formulation', 'discontinuous galerkin method'])\n", "('NONE', 'a consistent XXXXX of the nonlocal XXXXX is presented', ['variational formulation', 'timoshenko beam'])\n", "('method used for task', 'use of volumetric XXXXX and pressure enhancement strategies for low XXXXX', ['strain rate', 'order finite elements'])\n", "('method used for task', 'six XXXXX are provided to investigate the strengths of the method through the XXXXX', ['numerical examples', 'software application'])\n", "('NONE', 'regular grid insertion to achieve XXXXX in XXXXX', ['linear complexity', 'delaunay triangulation'])\n", "('method used for task', 'use of a XXXXX and a XXXXX at two different stages of the same computation', ['3d model', 'beam model'])\n", "('NONE', 'a spline XXXXX constructed in practical domain can be dealt with as a XXXXX', ['meshless method', 'element method'])\n", "('NONE', 'a XXXXX with commercial XXXXX is developed to figure out the constraint equations', ['numerical method', 'fe codes'])\n", "('NONE', 'this fe has no numerical pathology ( XXXXX , XXXXX , … )', ['shear locking', 'spurious modes'])\n", "('NONE', 'XXXXX of higher-order XXXXX at interfaces', ['numerical implementation', 'boundary conditions'])\n", "('NONE', 'XXXXX gradients are calculated using weighted XXXXX', ['meshless method', 'plastic strain'])\n", "('NONE', 'a selected number of quasi-static and dynamic XXXXX has been chosen to show the performance of the new XXXXX', ['numerical examples', 'contact formulation'])\n", "('method used for task', 'several XXXXX operators between meshes are studied for XXXXX and remeshing', ['ale formulation', 'field transfer'])\n", "('NONE', 'a formulation of a 2d XXXXX with parabolic through-the-thickness XXXXX was carried out', ['temperature distribution', 'heat transfer element'])\n", "('method used for task', 'three XXXXX were developed to simulate XXXXX in pre-tensioned concrete and validated against previous experiments', ['numerical models', 'prestress transfer'])\n", "('method used for task', 'a XXXXX was conducted on factors affecting the XXXXX', ['parametric study', 'prestress transfer'])\n", "('method used for task', 'a meshfree interface-finite XXXXX ( mi-fem ) for XXXXX', ['phase transformation', 'element method'])\n", "('method used for task', 'solving XXXXX for strain gradient XXXXX', ['mixed finite element formulation', 'elasticity problems'])\n", "('method used for task', 'a novel and simple fsdt-based XXXXX for XXXXX of fgm plates is presented', ['isogeometric analysis', 'geometrically nonlinear analysis'])\n", "('NONE', 'simplified XXXXX element formulation satisfying a priori the XXXXX', ['finite strain', 'patch test'])\n", "('method used for task', 'the proposed XXXXX is based on the XXXXX', ['optimization procedure', 'beso method'])\n", "('NONE', 'the objective is to minimize the XXXXX of the XXXXX', ['frequency response', 'coupled systems'])\n", "('NONE', 'XXXXX of topologically complex XXXXX is performed by using trimmed nurbs surfaces and exact normal vectors', ['isogeometric analysis', 'shell structures'])\n", "('NONE', 'XXXXX of topologically complex shell structures is performed by using trimmed XXXXX and exact normal vectors', ['isogeometric analysis', 'nurbs surfaces'])\n", "('method used for task', 'isogeometric analysis of topologically complex XXXXX is performed by using trimmed XXXXX and exact normal vectors', ['shell structures', 'nurbs surfaces'])\n", "('method used for task', 'the exact normal vectors , which are directly calculated by XXXXX are used in the shell formulation based on the reissner–mindlin degenerated XXXXX', ['shell element', 'nurbs surface expression'])\n", "('method used for task', 'the exact normal vectors , which are directly calculated by nurbs surface expression are used in the XXXXX based on the reissner–mindlin degenerated XXXXX', ['shell element', 'shell formulation'])\n", "('NONE', 'the exact normal vectors , which are directly calculated by XXXXX are used in the XXXXX based on the reissner–mindlin degenerated shell element', ['nurbs surface expression', 'shell formulation'])\n", "('NONE', 'the analytic derivatives of the direction vectors , which are directly calculated by XXXXX are also employed in the XXXXX', ['nurbs surface expression', 'shell formulation'])\n", "('method used for task', 'with XXXXX , it is shown that the present method for topologically complex XXXXX , which can not be treated by the conventional isogeometric analysis without multiple patches , gives appropriate solutions', ['numerical examples', 'shell structures'])\n", "('method used for task', 'with XXXXX , it is shown that the present method for topologically complex shell structures , which can not be treated by the conventional XXXXX without multiple patches , gives appropriate solutions', ['numerical examples', 'isogeometric analysis'])\n", "('NONE', 'with numerical examples , it is shown that the present method for topologically complex XXXXX , which can not be treated by the conventional XXXXX without multiple patches , gives appropriate solutions', ['shell structures', 'isogeometric analysis'])\n", "('NONE', 'a simple XXXXX for prediction of buckled XXXXX is proposed', ['analytical model', 'beam stiffness'])\n", "('NONE', 'XXXXX of the XXXXX is achieved using a model reduction technique', ['fast computation', 'wave modes'])\n", "('NONE', 'large XXXXX are obtained when computing the forced response of XXXXX', ['periodic structures', 'cpu time savings'])\n", "('NONE', 'established the effectiveness of the three-phase XXXXX approach in capturing the non-linearity of the materials considered without sacrificing the simplicity of the XXXXX model', ['unit cell', 'unit cell approach'])\n", "('NONE', 'different XXXXX were also conducted to identify the influence of certain phase characteristics of the XXXXX for the materials considered', ['parametric studies', 'unit cell'])\n", "('NONE', 'XXXXX of the multiscale XXXXX are derived on some regularity hypothesis', ['error estimates', 'approximate solution'])\n", "('NONE', 'XXXXX enables an XXXXX of fatigue stress–strain loop', ['efficient computation', 'control protocol'])\n", "('method used for task', 'XXXXX are successfully tested for XXXXX and structures', ['control algorithms', 'volume elements'])\n", "('method used for task', 'meshing procedures have a significant effect on the XXXXX and the von mises stress ( XXXXX vs. brick mesh )', ['tetrahedral mesh', 'fragmentation behavior'])\n", "('NONE', 'second-order effects are considered by XXXXX , save XXXXX', ['computational time', 'stability functions'])\n", "('NONE', 'first static and dynamic study of lsfes for rm XXXXX on XXXXX', ['composite plates', 'unstructured grids'])\n", "('NONE', 'first comparison of rm lsfes with shell281 elements in terms of XXXXX , XXXXX', ['cpu time', 'model size'])\n", "('method used for task', 'the proposed XXXXX is based on the XXXXX', ['optimization procedure', 'beso method'])\n", "('method used for task', 'the virtual-surface XXXXX , which is based on inside–outside method and penalty function method , is developed . innermost details regarding XXXXX , contact force calculation and the algorithmic implementations are presented', ['local search', 'contact algorithm'])\n", "('method used for task', 'in the proposed contact algorithm , the penetration between XXXXX and XXXXX can be determined without solving nonlinear equations', ['discrete element', 'finite element'])\n", "('NONE', 'in the proposed XXXXX , the penetration between XXXXX and finite element can be determined without solving nonlinear equations', ['discrete element', 'contact algorithm'])\n", "('method used for task', 'in the proposed XXXXX , the penetration between discrete element and XXXXX can be determined without solving nonlinear equations', ['finite element', 'contact algorithm'])\n", "('NONE', 'the proposed XXXXX has been implemented into in-house developed code . four XXXXX are employed to validate the accuracy and robustness of the proposed XXXXX', ['numerical examples', 'contact algorithm'])\n", "('NONE', 'the proposed XXXXX has been implemented into in-house developed code . four XXXXX are employed to validate the accuracy and robustness of the proposed XXXXX', ['numerical examples', 'contact algorithm'])\n", "('NONE', 'buckling and XXXXX of dimpled XXXXX were investigated', ['ultimate strength', 'steel columns'])\n", "('method used for task', 'extension of previous work of the authors on constitutive contact laws in combination with the dual mortar method for XXXXX to quadratic XXXXX', ['contact problems', 'finite elements'])\n", "('method used for task', 'suggestion of a petrov galerkin dual mortar approach for quadratic XXXXX to avoid XXXXX in case of large curvatures or high gradients', ['finite elements', 'consistency errors'])\n", "('method used for task', 'a mode patch that makes the algorithm applicable to both XXXXX and three-dimensional XXXXX', ['plane stress', 'stress states'])\n", "('NONE', 'the XXXXX permits the use of gradient-based methods in XXXXX in axisymmetry with torsional loads', ['sensitivity analysis', 'shape optimization'])\n", "('method used for task', 'an efficient XXXXX is developed for XXXXX of 3d bi-modulus materials', ['computational method', 'geometrically nonlinear analysis'])\n", "('method used for task', 'the proposed model provide high-precision XXXXX and local XXXXX', ['load distribution', 'stress field'])\n", "('method used for task', 'the new fe model considers both XXXXX and size effects for micro XXXXX of circular cups', ['surface roughness', 'deep drawing'])\n", "('NONE', 'the new XXXXX considers both XXXXX and size effects for micro deep drawing of circular cups', ['surface roughness', 'fe model'])\n", "('method used for task', 'the new XXXXX considers both surface roughness and size effects for micro XXXXX of circular cups', ['deep drawing', 'fe model'])\n", "('NONE', 'XXXXX affects the springback , the drawability and the cups’ quality obviously in micro XXXXX', ['surface roughness', 'deep drawing'])\n", "('method used for task', 'the XXXXX is given by a residual-stress dependent nonlinear elastic XXXXX in terms of invariants', ['material model', 'constitutive law'])\n", "('method used for task', 'the dependence of bifurcation and postbifurcation behavior of tubes under torsion on XXXXX is illustrated and compared with results when there is no XXXXX', ['residual stresses', 'residual stress'])\n", "('NONE', 'XXXXX of the coupled governing XXXXX', ['differential equations', 'solution strategy'])\n", "('NONE', 'a method is proposed to blend nurbs and XXXXX in 3d XXXXX', ['finite elements', 'isogeometric analysis'])\n", "('method used for task', 'a mesh-adaptive XXXXX has been implemented to work along with the non-local XXXXX', ['finite element technique', 'damage model'])\n", "('method used for task', 'development of a novel XXXXX for the combined finite-discrete XXXXX by means of ghost particles', ['coupling approach', 'element method'])\n", "('NONE', 'development of a novel XXXXX for the combined finite-discrete element method by means of XXXXX', ['coupling approach', 'ghost particles'])\n", "('NONE', 'development of a novel coupling approach for the combined finite-discrete XXXXX by means of XXXXX', ['element method', 'ghost particles'])\n", "('method used for task', 'we extend entropy and its relevance in XXXXX , e.g . microsate , ensemble , and information , to evaluate the XXXXX of elastic bodies , which are discretized with finite elements , some concepts in contact mechanics are considered as the same or similar meaning of that in XXXXX', ['statistical physics', 'contact forces'])\n", "('method used for task', 'we extend entropy and its relevance in XXXXX , e.g . microsate , ensemble , and information , to evaluate the contact forces of elastic bodies , which are discretized with XXXXX , some concepts in contact mechanics are considered as the same or similar meaning of that in XXXXX', ['statistical physics', 'finite elements'])\n", "('NONE', 'we extend entropy and its relevance in statistical physics , e.g . microsate , ensemble , and information , to evaluate the XXXXX of elastic bodies , which are discretized with XXXXX , some concepts in contact mechanics are considered as the same or similar meaning of that in statistical physics', ['contact forces', 'finite elements'])\n", "('method used for task', 'we extend entropy and its relevance in XXXXX , e.g . microsate , ensemble , and information , to evaluate the XXXXX of elastic bodies , which are discretized with finite elements , some concepts in contact mechanics are considered as the same or similar meaning of that in XXXXX', ['contact forces', 'statistical physics'])\n", "('method used for task', 'we extend entropy and its relevance in XXXXX , e.g . microsate , ensemble , and information , to evaluate the contact forces of elastic bodies , which are discretized with XXXXX , some concepts in contact mechanics are considered as the same or similar meaning of that in XXXXX', ['finite elements', 'statistical physics'])\n", "('method used for task', 'we obtain an explicit XXXXX for the normalized XXXXX by maximizing entropy subject to an expectation value and the principle of minimum potential energy', ['probability distribution', 'contact forces'])\n", "('method used for task', 'we obtain an explicit XXXXX for the normalized contact forces by maximizing entropy subject to an expectation value and the principle of minimum XXXXX', ['probability distribution', 'potential energy'])\n", "('method used for task', 'we obtain an explicit probability distribution for the normalized XXXXX by maximizing entropy subject to an expectation value and the principle of minimum XXXXX', ['contact forces', 'potential energy'])\n", "('NONE', 'we construct an efficient XXXXX by solving a series of isolated systems to obtain the XXXXX , and give a novelty termination criteria to terminate the iteration', ['iterative procedure', 'contact forces'])\n", "('NONE', 'implemented two-scale XXXXX as a XXXXX for a 3d an eight-node interface element', ['damage model', 'material model'])\n", "('NONE', 'we present a new approach to compute the XXXXX of solid XXXXX', ['mass matrix', 'finite elements'])\n", "('NONE', 'XXXXX focus on consistent XXXXX of 10-node tetrahedral element', ['numerical examples', 'mass matrix'])\n", "('NONE', 'we specify an approach for the XXXXX of XXXXX', ['dynamic reconfiguration', 'mobile applications'])\n", "('method used for task', 'our offline learning approach with an XXXXX outperforms existing methods for XXXXX', ['adaptive genetic algorithm', 'model calibration'])\n", "('method used for task', 'employ hybrid XXXXX to generate representative views for XXXXX', ['3d models', 'feature lines'])\n", "('method used for task', 'we propose a XXXXX to efficiently solve the XXXXX', ['iterative algorithm', 'non-linear model'])\n", "('NONE', 'we define the well-posedness of XXXXX of the rational XXXXX', ['control polygon', 'bezier curve'])\n", "('method used for task', 'the XXXXX is injective if and only if its XXXXX is well-posed', ['bezier curve', 'control polygon'])\n", "('NONE', 'we present a XXXXX to determine the injectivity of the XXXXX', ['geometric method', 'bezier curve'])\n", "('NONE', 'geometrically XXXXX can be recognized by anatomists as meaningful locations on 3d XXXXX scans', ['salient points', 'human body'])\n", "('method used for task', 'efficient XXXXX for XXXXX is presented', ['compression algorithm', 'triangle meshes'])\n", "('method used for task', 'a method for identifying potential XXXXX that are subordinate to different XXXXX is present', ['feature points', 'feature lines'])\n", "('NONE', 'a feature selection framework is proposed to achieve XXXXX model-free XXXXX', ['high performance', 'gait recognition'])\n", "('method used for task', 'a XXXXX is proposed to achieve XXXXX model-free gait recognition', ['high performance', 'feature selection framework'])\n", "('method used for task', 'a XXXXX is proposed to achieve high performance model-free XXXXX', ['gait recognition', 'feature selection framework'])\n", "('method used for task', 'novel XXXXX based visual ego-motion estimation algorithm robust to abrupt XXXXX', ['particle filtering', 'camera motion'])\n", "('method used for task', 'multi-kernel XXXXX is a new facial XXXXX', ['appearance model', 'point detector'])\n", "('NONE', 'the XXXXX of the simplest XXXXX is barely increased', ['computational cost', 'modeling strategy'])\n", "('method used for task', 'introducing a XXXXX for simulating XXXXX', ['stochastic model', 'eye movements'])\n", "('method used for task', 'it extends the XXXXX to incorporate color and XXXXX', ['fast marching method', 'depth information'])\n", "('NONE', 'no strong constraints are imposed on the XXXXX or the XXXXX', ['camera motion', 'target appearance'])\n", "('method used for task', 'a XXXXX ( gp ) approach is proposed to design a combined XXXXX for image segmentation algorithms', ['genetic programming', 'evaluation measure'])\n", "('method used for task', 'a XXXXX ( gp ) approach is proposed to design a combined evaluation measure for XXXXX', ['genetic programming', 'image segmentation algorithms'])\n", "('method used for task', 'a genetic programming ( gp ) approach is proposed to design a combined XXXXX for XXXXX', ['evaluation measure', 'image segmentation algorithms'])\n", "('method used for task', 'we proposed a novel XXXXX based model for inhomogeneous XXXXX', ['level set', 'image segmentation'])\n", "('NONE', 'the local XXXXX can significantly increase XXXXX', ['image information', 'image contrast'])\n", "('method used for task', 'examined nine well-known models of XXXXX using the best metric to identify the best XXXXX models', ['visual saliency', 'visual saliency models'])\n", "('NONE', 'use of novel XXXXX with stereo features and XXXXX to disparity', ['appearance model', 'plane fitting'])\n", "('NONE', 'facade segmentation by XXXXX : hierarchical XXXXX', ['graphical model', 'markov random field'])\n", "('method used for task', 'allows the accurate regression of XXXXX to be scaled to XXXXX', ['gaussian processes', 'large data'])\n", "('NONE', 'we propose two strategies for XXXXX through XXXXX', ['face recognition', 'multiple features'])\n", "('method used for task', 'the XXXXX and XXXXX are learned simultaneously', ['dictionary learning', 'fusion process'])\n", "('method used for task', 'any XXXXX can be used , and errors are calculated on the original XXXXX', ['camera model', 'image space'])\n", "('NONE', 'the estimation starts with a XXXXX and ends with a XXXXX', ['random search', 'continuous optimization'])\n", "('NONE', 'XXXXX taking into account the XXXXX is introduced', ['distance function', 'hierarchical structure'])\n", "('method used for task', 'we propose an XXXXX for tracking XXXXX , capable of handling anomalies', ['online method', 'dense crowds'])\n", "('NONE', 'we propose a XXXXX of the problem based on XXXXX', ['nonlinear optimization', 'mutual information'])\n", "('NONE', 'we use XXXXX to find the scene that best renders the XXXXX', ['nonlinear optimization', 'input image'])\n", "('NONE', 'three XXXXX over two XXXXX are employed to support the conclusions', ['statistical tests', 'performance measures'])\n", "('NONE', 'XXXXX was implemented using bayes XXXXX', ['model comparison', 'information criterion'])\n", "('method used for task', 'XXXXX is incorporated by employing XXXXX into clips', ['motion analysis', 'motion segmentation'])\n", "('method used for task', 'we propose an ensemble XXXXX ( edl ) framework for XXXXX', ['dictionary learning', 'saliency detection'])\n", "('method used for task', 'XXXXX based method for XXXXX using multiple views', ['data fusion', 'activity recognition'])\n", "('method used for task', 'XXXXX based method for activity recognition using XXXXX', ['data fusion', 'multiple views'])\n", "('NONE', 'data fusion based method for XXXXX using XXXXX', ['activity recognition', 'multiple views'])\n", "('NONE', 'we evaluate the XXXXX in XXXXX', ['object recognition', '3d shape descriptor'])\n", "('NONE', 'eyebrow gestures and periodic XXXXX convey XXXXX', ['linguistic information', 'head movements'])\n", "('NONE', 'our methods can enhance XXXXX lying in the low frequency part of XXXXX', ['face features', 'face images'])\n", "('NONE', 'a statistical XXXXX yields the complexity of the XXXXX', ['stability analysis', 'registration model'])\n", "('NONE', 'we reduce the XXXXX of fingertip locations conditioned to a XXXXX', ['search space', 'hand gesture'])\n", "('method used for task', 'XXXXX of 99 % ( almost perfect ) is reached to detect XXXXX', ['classification accuracy', 'mask attacks'])\n", "('method used for task', 'XXXXX tackling speaker dependency , head poses and XXXXX', ['visual features', 'temporal information'])\n", "('NONE', 'for XXXXX , a hierarchical piece-wise XXXXX is examined', ['non-rigid registration', 'affine transform'])\n", "('method used for task', \"we examine XXXXX on the proposed metric and moran 's XXXXX\", ['roc analysis', 'spatial correlation'])\n", "('method used for task', 'XXXXX is used to combine various XXXXX and determine a safety score', ['fuzzy logic', 'attribute values'])\n", "('NONE', 'XXXXX provides a XXXXX for textural images in each class', ['sparse representation', 'compact model'])\n", "('NONE', 'distributions of gmrf local XXXXX as improved XXXXX', ['texture descriptors', 'parameter estimates'])\n", "('method used for task', 'we propose a novel XXXXX for XXXXX', ['action recognition', 'deep learning model'])\n", "('method used for task', 'an integration of XXXXX , machine listening , and XXXXX', ['computer vision', 'machine learning'])\n", "('NONE', 'we introduce the 3dspmk for object and XXXXX in XXXXX', ['scene recognition', 'depth images'])\n", "('method used for task', 'we design a scale-invariant XXXXX for shape matching and XXXXX', ['shape descriptor', 'object detection'])\n", "('NONE', 'the proposed tracker combines the flexibility of XXXXX and robustness of XXXXX', ['interest points', 'sparse representation'])\n", "('method used for task', 'a directed XXXXX is proposed for XXXXX', ['structural model', 'feature correspondence'])\n", "('method used for task', 'we propose two novel XXXXX for unsupervised non-gaussian XXXXX', ['feature selection', 'inference frameworks'])\n", "('method used for task', 'a regional bounding spherical descriptor is used for XXXXX and XXXXX', ['3d face', 'emotion analysis'])\n", "('method used for task', 'dst-klpp provides higher XXXXX than other methods , including wavelet , curvelet and XXXXX', ['contourlet transform', 'classification rates'])\n", "('method used for task', 'the proposed model gives a XXXXX and a good balance between XXXXX and processing efficiency', ['real-time performance', 'segmentation accuracy'])\n", "('method used for task', 'a study of the use of XXXXX for XXXXX is presented', ['topic models', 'video retrieval'])\n", "('NONE', 'the results highlight the XXXXX among the XXXXX', ['topic models', 'performance differences'])\n", "('method used for task', 'a XXXXX has been introduced in the process to reduce the XXXXX of particle filtering', ['markov chain model', 'computational complexity'])\n", "('NONE', 'a XXXXX has been introduced in the process to reduce the computational complexity of XXXXX', ['markov chain model', 'particle filtering'])\n", "('NONE', 'a markov chain model has been introduced in the process to reduce the XXXXX of XXXXX', ['computational complexity', 'particle filtering'])\n", "('NONE', 'application of XXXXX along with a XXXXX to refine any outliers', ['ransac algorithm', 'histogram technique'])\n", "('method used for task', 'hierarchical XXXXX is proposed to obtain an acceptable solution in XXXXX', ['real time', 'diffusion method'])\n", "('method used for task', 'XXXXX and XXXXX between objects and shadows are used to recover misdetected shadows', ['mutual information', 'data association'])\n", "('method used for task', 'a effective XXXXX is introduced to optimize the XXXXX', ['iterative algorithm', 'objective function'])\n", "('NONE', 'new multi-lateral filter to efficiently increase the XXXXX of low-resolution and noisy XXXXX in real-time', ['spatial resolution', 'depth maps'])\n", "('NONE', 'the proposed filter has been effectively and efficiently implemented for XXXXX in XXXXX', ['dynamic scenes', 'real-time applications'])\n", "('NONE', 'we examine the role of XXXXX and image semantics in understanding XXXXX', ['visual attention', 'image memorability'])\n", "('method used for task', 'we propose an attention-driven spatial XXXXX for XXXXX', ['pooling strategy', 'image memorability'])\n", "('NONE', 'an algorithm has been developed for XXXXX of XXXXX', ['automatic detection', 'vanishing points'])\n", "('method used for task', 'the results have been analyzed according to the number of XXXXX and XXXXX', ['vanishing points', 'image resolution'])\n", "('method used for task', 'XXXXX is done by the pre-trained svm using 7 different XXXXX', ['state transition', 'input features'])\n", "('method used for task', 'we also learn XXXXX from very few positive and related XXXXX', ['event detectors', 'training samples'])\n", "('method used for task', 'in the XXXXX each region gives rise to a learnt weak XXXXX', ['ranking function', 'training step'])\n", "('method used for task', 'proposed XXXXX for XXXXX', ['ensemble learning', 'face hallucination'])\n", "('method used for task', 'designed a pre-evaluation stage to save XXXXX and reduce XXXXX', ['computation time', 'false alarm rate'])\n", "('method used for task', 'symstereo and pearl are combined for carrying the XXXXX and XXXXX simultaneously', ['3d reconstruction', 'plane fitting'])\n", "('NONE', 'a new dual many-to-one encoder method for XXXXX across XXXXX', ['feature extraction', 'action datasets'])\n", "('method used for task', 'no training or XXXXX is needed and achieve XXXXX', ['camera calibration', 'real-time processing'])\n", "('method used for task', 'we propose a new local XXXXX for XXXXX', ['action recognition', 'part model'])\n", "('method used for task', 'XXXXX and fast XXXXX are achieved', ['high performance', 'action recognition'])\n", "('NONE', 'a 3-axis gyro mounted to a XXXXX can frequently predict the main component of XXXXX', ['video camera', 'optical flow'])\n", "('method used for task', 'gyro regularization adds very little XXXXX to XXXXX', ['computational cost', 'feature tracking'])\n", "('NONE', 'the new XXXXX improves results in complex XXXXX', ['kernel framework', 'activities recognition'])\n", "('method used for task', 'we apply collective XXXXX to cross-domain XXXXX', ['matrix factorization', 'action recognition'])\n", "('NONE', 'XXXXX learned by the structured output XXXXX', ['support vector machines', 'landmark detector'])\n", "('method used for task', 'we simultaneously learn XXXXX and XXXXX', ['visual features', 'hash functions'])\n", "('method used for task', 'a timely review of XXXXX based on spatio-temporal XXXXX', ['action recognition', 'local features'])\n", "('method used for task', 'transfer techniques in XXXXX to video domain for XXXXX', ['action recognition', 'image domain'])\n", "('method used for task', 'handle the misalignment problems in XXXXX and XXXXX', ['image classification', 'action recognition'])\n", "('NONE', 'improve the discrimination of XXXXX in XXXXX', ['feature pooling', 'visual recognition tasks'])\n", "('NONE', 'uses the notion of XXXXX and a pattern comic to make sense of the constitutive ‘mechanics’ of XXXXX', ['data structures', 'business patterns'])\n", "('NONE', 'grounds the application of XXXXX and pattern comics in material collected within an XXXXX study', ['action research', 'business patterns'])\n", "('method used for task', 'we propose a link-bridged XXXXX for cross-domain XXXXX', ['topic model', 'document classification'])\n", "('NONE', 'directions for future works related to XXXXX in XXXXX', ['metadata quality', 'data infrastructures'])\n", "('NONE', 'we present an XXXXX for the study of cross-lingual XXXXX', ['evaluation framework', 'link discovery'])\n", "('NONE', 'XXXXX contain XXXXX and evolve over time', ['xml documents', 'temporal information'])\n", "('NONE', 'our work explores XXXXX , versioning and querying support in XXXXX', ['change detection', 'xml documents'])\n", "('NONE', 'we investigate various query XXXXX using XXXXX', ['estimation methods', 'bias–variance analysis'])\n", "('NONE', 'we propose a new method for XXXXX validity using cross XXXXX', ['outlier detection', 'term translation'])\n", "('method used for task', 'we show that variance of rn sampling grows with XXXXX for XXXXX', ['scale-free networks', 'data size'])\n", "('NONE', 'we model bid XXXXX as a mixed XXXXX', ['keyword suggestion', 'integer optimization problem'])\n", "('NONE', 'enriching queries using XXXXX increases XXXXX in healthcare domain', ['user preferences', 'information retrieval'])\n", "('NONE', 'prioritizing XXXXX improves XXXXX', ['user preferences', 'service discovery'])\n", "('method used for task', 'we combine active and XXXXX for cross-lingual XXXXX', ['semi-supervised learning', 'sentiment classification'])\n", "('method used for task', 'density analysis of XXXXX is used in XXXXX', ['unlabeled data', 'active learning'])\n", "('NONE', 'XXXXX of XXXXX is used in active learning', ['unlabeled data', 'density analysis'])\n", "('NONE', 'XXXXX of unlabeled data is used in XXXXX', ['active learning', 'density analysis'])\n", "('NONE', 'results show that incorporating XXXXX can speed up XXXXX', ['learning process', 'density analysis'])\n", "('NONE', 'we implement an active learning scenario for XXXXX on XXXXX', ['sentiment analysis', 'data streams'])\n", "('method used for task', 'XXXXX are shown to be suitable for XXXXX', ['sentiment analysis', 'machine learning methods'])\n", "('NONE', 'XXXXX improves the accuracy of XXXXX', ['active learning', 'sentiment classification'])\n", "('method used for task', 'the use of wikipedia as a XXXXX for XXXXX', ['knowledge source', 'question answering system'])\n", "('NONE', 'experimental results indicate XXXXX in term of XXXXX', ['high accuracy', 'bleu score'])\n", "('NONE', 'we explore state-of-the-art supervised machine learning methods for XXXXX of czech XXXXX', ['sentiment analysis', 'social media'])\n", "('method used for task', 'we explore state-of-the-art XXXXX for XXXXX of czech social media', ['sentiment analysis', 'supervised machine learning methods'])\n", "('NONE', 'we explore state-of-the-art XXXXX for sentiment analysis of czech XXXXX', ['social media', 'supervised machine learning methods'])\n", "('method used for task', 'combine grouping of video search results with XXXXX to assist XXXXX', ['video retrieval', 'recommendation techniques'])\n", "('NONE', 'we carry out an XXXXX to determine characteristics of XXXXX', ['empirical analysis', 'social media channels'])\n", "('method used for task', 'a new XXXXX for the next-generation XXXXX', ['theoretical framework', 'recommender systems'])\n", "('NONE', 'XXXXX of learning couplings in XXXXX and recommendation', ['case studies', 'data mining'])\n", "('NONE', 'the rise of XXXXX has fueled interest in XXXXX', ['social media', 'sentiment classification'])\n", "('method used for task', 'relationship between XXXXX and XXXXX : common and different reasons among different knowledge groups', ['task difficulty', 'user knowledge'])\n", "('NONE', 'we tackle the problem of XXXXX differently , by XXXXX', ['data fusion', 'microblog search'])\n", "('NONE', 'tuning XXXXX increases XXXXX', ['genetic algorithm', 'time efficiency'])\n", "('NONE', 'expressive signals enrich the XXXXX of baseline and XXXXX', ['feature space', 'ensemble classifiers'])\n", "('method used for task', 'means uses XXXXX and standards for XXXXX and integration', ['semantic web technologies', 'data sharing'])\n", "('NONE', 'we carry out an XXXXX to determine characteristics of XXXXX', ['empirical analysis', 'social media channels'])\n", "('NONE', 'we explore state-of-the-art supervised machine learning methods for XXXXX of czech XXXXX', ['sentiment analysis', 'social media'])\n", "('method used for task', 'we explore state-of-the-art XXXXX for XXXXX of czech social media', ['sentiment analysis', 'supervised machine learning methods'])\n", "('NONE', 'we explore state-of-the-art XXXXX for sentiment analysis of czech XXXXX', ['social media', 'supervised machine learning methods'])\n", "('NONE', 'we face the XXXXX of having a limited set of XXXXX', ['pairwise constraints', 'real-world problem'])\n", "('method used for task', 'show effectiveness with above 70 % XXXXX for user XXXXX', ['prediction accuracy', 'search performance'])\n", "('method used for task', 'XXXXX effectively employed to improve XXXXX', ['unlabeled data', 'classification performance'])\n", "('NONE', 'we build a XXXXX that can be tested by the XXXXX', ['prototype system', 'research community'])\n", "('NONE', 'improving XXXXX by integrating co-occurrence relations into XXXXX', ['translation quality', 'word models'])\n", "('NONE', 'comparing different estimations of XXXXX from XXXXX', ['translation probabilities', 'word correlations'])\n", "('NONE', 'we perform XXXXX of personality , XXXXX and mood sharing in twitter', ['correlation analysis', 'communication style'])\n", "('method used for task', 'XXXXX ( gp ) algorithm has been employed for XXXXX', ['genetic programming', 'feature learning'])\n", "('method used for task', 'we present a new XXXXX for relations between concepts based on XXXXX of concepts', ['weighting scheme', 'distributed representations'])\n", "('NONE', 'examine the XXXXX of XXXXX on gens y and z', ['the internet', 'influence factors'])\n", "('NONE', 'we study the characteristics of the XXXXX ( sk ) of XXXXX', ['straight skeleton', 'monotone polygons'])\n", "('method used for task', 'we propose a XXXXX and a XXXXX for social collaboration processes', ['modeling approach', 'visual modeling notation'])\n", "('NONE', 'framework of ten XXXXX functionalities ( cmf ) in XXXXX is defined', ['business processes', 'compliance monitoring'])\n", "('method used for task', 'XXXXX and presentation of real-world business constraints for compliance are extracted from five XXXXX', ['systematic literature review', 'case studies'])\n", "('NONE', 'we consider the comparison with rtl power estimation techniques on several design aspects : XXXXX , XXXXX and modeling effort ,', ['estimation accuracy', 'simulation time'])\n", "('NONE', 'XXXXX appropriate for predictive maintenance of XXXXX', ['mathematical model', 'transmission line'])\n", "('NONE', 'distributed XXXXX and improvement of delivered XXXXX', ['power quality', 'energy resources'])\n", "('NONE', 'examines a XXXXX of normative prototypes that reshape the issue of XXXXX', ['case study', 'aircraft noise'])\n", "('NONE', 'XXXXX avoids XXXXX in ihe xds shared ehr systems', ['content-based search', 'information overload'])\n", "('method used for task', 'XXXXX can be used to structure information from XXXXX reports', ['natural language processing', 'free text'])\n", "('method used for task', 'XXXXX can be used to XXXXX from free text reports', ['natural language processing', 'structure information'])\n", "('NONE', 'natural language processing can be used to XXXXX from XXXXX reports', ['free text', 'structure information'])\n", "('method used for task', 'XXXXX approaches are yet to be fully explored in XXXXX of cancer information', ['machine learning', 'text mining'])\n", "('method used for task', 'XXXXX are feasible platforms for self-supervised XXXXX', ['mobile devices', 'cognitive training'])\n", "('method used for task', 'self-guided and internet-delivered cognitive XXXXX ( icbt ) has potential for XXXXX', ['older adults', 'behavior therapy'])\n", "('NONE', 'the study result show that adults with adhd benefit from a coach-guided XXXXX using smartphone applications and that XXXXX are a feasible way to reach this patient group', ['internet interventions', 'online intervention'])\n", "('NONE', 'the study result show that adults with adhd benefit from a coach-guided online intervention using XXXXX and that XXXXX are a feasible way to reach this patient group', ['internet interventions', 'smartphone applications'])\n", "('NONE', 'the study result show that adults with adhd benefit from a coach-guided XXXXX using XXXXX and that internet interventions are a feasible way to reach this patient group', ['online intervention', 'smartphone applications'])\n", "('method used for task', 'the effectiveness of a guided and unguided act-based XXXXX for XXXXX ( actonpain ) will be investigated', ['chronic pain', 'online intervention'])\n", "('NONE', 'cognitive behavioural therapy via XXXXX ( icbt ) may benefit XXXXX', ['the internet', 'university students'])\n", "('NONE', 'a XXXXX of client e-mails was conducted within the setting of a guided internet-based cognitive XXXXX program for depression', ['content analysis', 'behavior therapy'])\n", "('method used for task', 'XXXXX is a depleting XXXXX , however the availability of effective treatments are scarce', ['chronic pain', 'health problem'])\n", "('method used for task', 'the presented method provides a XXXXX as basis for XXXXX', ['bayesian framework', 'data fusion'])\n", "('method used for task', 'XXXXX is interpreted by a XXXXX for copd-exacerbation detection', ['patient data', 'bayesian network model'])\n", "('method used for task', 'explorer provides high XXXXX and strong XXXXX', ['estimation accuracy', 'privacy protection'])\n", "('NONE', 'automatic pgx-specific drug–gene XXXXX from XXXXX is difficult', ['free text', 'relationship extraction'])\n", "('NONE', 'we develop a semi-supervised XXXXX method requiring minimal prior XXXXX', ['domain knowledge', 'relationship extraction'])\n", "('method used for task', 'we created software integrated model for XXXXX and XXXXX ( impact )', ['patient care', 'clinical trials'])\n", "('NONE', 'impact coordinates XXXXX visits with XXXXX visits', ['clinical research', 'patient care'])\n", "('NONE', 'examines health outcomes , effects and affordances of XXXXX in XXXXX', ['social media', 'chronic disease'])\n", "('method used for task', 'XXXXX is needed in XXXXX of gene expression', ['gene selection', 'supervised classification'])\n", "('NONE', 'XXXXX is needed in supervised classification of XXXXX', ['gene selection', 'gene expression'])\n", "('NONE', 'gene selection is needed in XXXXX of XXXXX', ['supervised classification', 'gene expression'])\n", "('NONE', 'XXXXX of electronic XXXXX is achieved for safety studies', ['secondary use', 'health records'])\n", "('method used for task', 'XXXXX of electronic health records is achieved for XXXXX', ['secondary use', 'safety studies'])\n", "('method used for task', 'secondary use of electronic XXXXX is achieved for XXXXX', ['health records', 'safety studies'])\n", "('method used for task', 'combining XXXXX and XXXXX , consensus will be fostered', ['clinical decision support', 'social network'])\n", "('NONE', 'XXXXX with the XXXXX encourages adoption and collaboration', ['open source', 'social network'])\n", "('method used for task', 'we review the use of XXXXX for health applied to XXXXX', ['mobile phones', 'older adults'])\n", "('NONE', 'the XXXXX was tested on two XXXXX using 2150 production radiology reports', ['similarity measure', 'classification problems'])\n", "('NONE', 'XXXXX can exploit the vast amounts of XXXXX stored in emrs', ['semi-supervised learning', 'unlabeled data'])\n", "('method used for task', 'XXXXX and XXXXX technologies can be used to deal with biology’s XXXXX sets', ['cloud computing', 'big data'])\n", "('method used for task', 'XXXXX and XXXXX can be used to deal with biology’s big data sets', ['cloud computing', 'big data technologies'])\n", "('method used for task', 'XXXXX and big data technologies can be used to deal with biology’s XXXXX', ['cloud computing', 'big data sets'])\n", "('method used for task', 'cloud computing and XXXXX technologies can be used to deal with biology’s XXXXX sets', ['big data', 'big data technologies'])\n", "('method used for task', 'cloud computing and XXXXX technologies can be used to deal with biology’s XXXXX sets', ['big data', 'big data sets'])\n", "('method used for task', 'cloud computing and XXXXX can be used to deal with biology’s XXXXX', ['big data technologies', 'big data sets'])\n", "('method used for task', 'challenges associated with XXXXX and XXXXX in biology are discussed', ['cloud computing', 'big data technologies'])\n", "('NONE', 'three different models for inferring XXXXX from XXXXX are proposed', ['gene networks', 'microarray data'])\n", "('NONE', 'semantator converts biomedical text to XXXXX with a XXXXX', ['linked data', 'formal semantics'])\n", "('method used for task', 'we develop a new XXXXX and query-document XXXXX', ['vector space model', 'similarity function'])\n", "('NONE', 'successfully used as XXXXX in the semeval-2013 ddi XXXXX', ['gold standard', 'extraction task'])\n", "('method used for task', 'a dynamic tag cloud may reduce XXXXX for XXXXX', ['information overload', 'clinical trial search'])\n", "('method used for task', 'we investigated how XXXXX and XXXXX changed over time', ['user interaction', 'usability problems'])\n", "('method used for task', 'we propose a XXXXX to discover underlying patterns in XXXXX', ['probabilistic model', 'clinical pathways'])\n", "('NONE', 'XXXXX of XXXXX and mappings evolution', ['quantitative analysis', 'medical ontologies'])\n", "('method used for task', 'we proposed a XXXXX ( graph-based and machine-learning ) for temporal XXXXX', ['hybrid system', 'relation extraction'])\n", "('NONE', 'XXXXX shows superiority of our XXXXX on our utility measures', ['empirical evaluation', 'model based'])\n", "('NONE', 'XXXXX can support XXXXX and analysis', ['classification systems', 'data retrieval'])\n", "('NONE', 'discovery of XXXXX in XXXXX using literature knowledge', ['drug–drug interactions', 'patient data'])\n", "('NONE', 'potential XXXXX identified in medication lists from XXXXX', ['drug–drug interactions', 'clinical data'])\n", "('method used for task', 'XXXXX with globus transfer for high-performance and reliable XXXXX', ['data transfer', 'integrate galaxy'])\n", "('method used for task', 'XXXXX with htcondor scheduler for auto-scaling and XXXXX', ['parallel computing', 'integrate galaxy'])\n", "('method used for task', 'two bioinformatics workflow XXXXX and XXXXX are presented', ['use cases', 'performance evaluation'])\n", "('NONE', 'our approach combines XXXXX visual queries , mining , and XXXXX', ['ad hoc', 'interactive visualization'])\n", "('NONE', 'a feature-based approach identifies XXXXX with similar XXXXX', ['clinical trials', 'eligibility text'])\n", "('method used for task', 'we propose a XXXXX for detecting XXXXX in mammograms', ['cad system', 'breast cancer'])\n", "('NONE', 'XXXXX optimized XXXXX detects the cancers', ['swarm intelligence', 'wavelet neural network'])\n", "('method used for task', 'we focus on optimized XXXXX to enhance the XXXXX', ['wavelet neural network', 'detection accuracy'])\n", "('NONE', 'disorders , findings , drugs and XXXXX were annotated in swedish XXXXX', ['clinical text', 'body parts'])\n", "('method used for task', 'we propose a new prognostic XXXXX based on a bayesian XXXXX', ['prediction model', 'evolutionary learning'])\n", "('NONE', 'the proposed bayesian XXXXX effectively handles a complex XXXXX', ['evolutionary learning', 'search space'])\n", "('NONE', 'the performance of XXXXX outperforms other XXXXX', ['classification models', 'hypergraph classifiers'])\n", "('NONE', 'XXXXX of XXXXX for research is a complex task but it benefits the entire medical center community', ['secondary use', 'clinical data'])\n", "('method used for task', 'change in XXXXX and XXXXX were integrated simultaneously', ['gene expression', 'structural information'])\n", "('NONE', 'XXXXX include XXXXX , task accuracy , and path length', ['performance metrics', 'completion time'])\n", "('method used for task', 'XXXXX include completion time , task accuracy , and XXXXX', ['performance metrics', 'path length'])\n", "('method used for task', 'performance metrics include XXXXX , task accuracy , and XXXXX', ['completion time', 'path length'])\n", "('NONE', 'how XXXXX varies with accuracy index and XXXXX', ['sample size', 'effect size'])\n", "('method used for task', 'XXXXX is gathered by a wearable sensor and sent to a XXXXX', ['mobile device', 'ecg data'])\n", "('NONE', 'we propose an XXXXX that considers the semantic of XXXXX', ['medical images', 'image retrieval system'])\n", "('method used for task', 'XXXXX is used to search for a good XXXXX', ['evolutionary programming', 'discretization scheme'])\n", "('NONE', 'XXXXX observed in colposcopy are used as predictors of XXXXX', ['temporal patterns', 'cervical cancer'])\n", "('NONE', 'propose XXXXX that allow XXXXX to make collective decisions', ['ensemble methods', 'base classifiers'])\n", "('NONE', 'two strategies enhanced the XXXXX ( dt ) proteins prediction power of XXXXX', ['drug target', 'svm classifier'])\n", "('method used for task', 'demonstrated utility of an in-domain collection ( XXXXX ) for XXXXX', ['clinical text', 'query expansion'])\n", "('NONE', 'the release of electronic XXXXX in the form of XXXXX may lead to privacy violations', ['health data', 'data streams'])\n", "('method used for task', 'XXXXX are shaped by the healthcare process and patient XXXXX', ['measurement patterns', 'health states'])\n", "('method used for task', 'XXXXX : the probability for suffering XXXXX in encrypted form', ['case study', 'cardiovascular disease'])\n", "('NONE', 'challenges include XXXXX , access , and limits of XXXXX and technology', ['data quality', 'human cognition'])\n", "('method used for task', 'we analyzed XXXXX and latent XXXXX for agreement on ctg evaluations', ['majority voting', 'class model'])\n", "('NONE', 'we built a novel architecture for XXXXX of XXXXX', ['information retrieval', 'patient records'])\n", "('NONE', 'annotating raw XXXXX generated the highest XXXXX', ['clinical text', 'quality data'])\n", "('NONE', 'XXXXX in electronic XXXXX ( ehr ) may re-identify patients', ['diagnosis codes', 'health records'])\n", "('NONE', 'we study discrete medical XXXXX with XXXXX', ['multi-label classification', 'time series datasets'])\n", "('method used for task', 'we present 3dfd.ujaen.es , a XXXXX for computing and analyzing the 3d XXXXX ( 3dfd )', ['web platform', 'fractal dimension'])\n", "('method used for task', 'the first XXXXX that allows the users to calculate , visualize , analyze and compare the 3dfd from XXXXX ( i.e . mri )', ['web platform', '3d images'])\n", "('NONE', 'we measure cds impact on XXXXX using statistical and XXXXX', ['cancer screening', 'simulation models'])\n", "('NONE', 'systematic studies of drug side XXXXX can facilitate XXXXX', ['drug discovery', 'effect associations'])\n", "('NONE', 'there is a need for techniques for XXXXX in medical XXXXX with events', ['data mining', 'time series'])\n", "('method used for task', 'XXXXX are automatically generated based on a XXXXX', ['regular expressions', 'domain ontology'])\n", "('NONE', 'first XXXXX of XXXXX in the biomedical domain', ['systematic review', 'text summarization'])\n", "('NONE', 'resolving the complexity of XXXXX from entity–attribute–value XXXXX', ['data extraction', 'data model'])\n", "('NONE', 'XXXXX may help biomedical researchers manage XXXXX', ['automatic summarization', 'information overload'])\n", "('NONE', 'ontological integration of XXXXX elements can compensate for deficiencies in XXXXX', ['data quality', 'ehr data'])\n", "('method used for task', 'the proposed XXXXX is achieved in the link layer without XXXXX', ['routing algorithm', 'route discovery'])\n", "('NONE', 'riig facilitates finding effective XXXXX in a XXXXX setting', ['drug targets', 'personalized medicine'])\n", "('NONE', 'riig bridges curated biomedical knowledge and XXXXX with XXXXX', ['causal reasoning', 'clinical data'])\n", "('NONE', 'we found evidence that the emr XXXXX are outpacing XXXXX', ['adoption changes', 'informatics research'])\n", "('method used for task', 'we propose an effective XXXXX for inferring XXXXX with priors', ['hybrid method', 'bayesian networks'])\n", "('NONE', 'the multi-technique approach includes XXXXX , XXXXX and ontology-based knowledgebase', ['information retrieval', 'natural language processing'])\n", "('NONE', 'two XXXXX are integrated in saps : XXXXX and pattern search', ['optimization methods', 'simulated annealing'])\n", "('method used for task', 'two XXXXX are integrated in saps : simulated annealing and XXXXX', ['optimization methods', 'pattern search'])\n", "('method used for task', 'two optimization methods are integrated in saps : XXXXX and XXXXX', ['simulated annealing', 'pattern search'])\n", "('NONE', 'XXXXX , such as visibility and XXXXX , is taken into account', ['prior knowledge', 'spatial constraints'])\n", "('method used for task', 'XXXXX ( color ) and XXXXX ( radius ) are depicted in circles', ['effect size', 'statistical significance'])\n", "('method used for task', 'we used XXXXX to model relations between outbreak and algorithm characteristics and XXXXX', ['bayesian networks', 'detection performance'])\n", "('NONE', 'we show that using nlp-based XXXXX improves adr XXXXX', ['feature extraction', 'classification accuracy'])\n", "('method used for task', 'a compilation of useful XXXXX for XXXXX is presented', ['bioinformatics tools', 'vaccine development'])\n", "('method used for task', 'we devise a XXXXX based algorithm on the reliable XXXXX', ['random walk', 'heterogeneous network'])\n", "('NONE', 'we discuss methods to use XXXXX in XXXXX for predictive modeling', ['temporal information', 'ehr data'])\n", "('NONE', 'our method was able to use XXXXX in XXXXX to improve performance', ['temporal information', 'ehr data'])\n", "('method used for task', 'XXXXX was used to estimate XXXXX of predictive features', ['multiple imputation', 'missing values'])\n", "('method used for task', 'XXXXX and XXXXX exploit information extrapolated from the unified medical language system', ['semantic similarity', 'relatedness measures'])\n", "('method used for task', 'evaluates combining XXXXX and XXXXX', ['semantic similarity', 'relatedness measures'])\n", "('NONE', 'we describe a method to generate word-concept XXXXX from a XXXXX', ['statistical models', 'knowledge base'])\n", "('method used for task', 'usage of XXXXX to create a XXXXX of the therapy process', ['hidden markov models', 'stochastic model'])\n", "('method used for task', 'the design and use of a XXXXX meter is analyzed using XXXXX', ['blood glucose', 'distributed cognition'])\n", "('NONE', 'we construct multiple XXXXX to make full use of all classes of XXXXX not only target samples', ['svdd models', 'training samples'])\n", "('method used for task', 'new method to determine XXXXX for building logistic XXXXX', ['sample size', 'prediction model'])\n", "('method used for task', 'XXXXX and aggregation techniques facilitate XXXXX', ['semantic similarity', 'hierarchical clustering'])\n", "('NONE', 'we constructed disease-related XXXXX using literature and XXXXX', ['gene network', 'google data'])\n", "('NONE', 'improved performance of the prediction system is reported using the XXXXX obtained through XXXXX', ['optimal threshold', 'pso algorithm'])\n", "('NONE', 'the extracted XXXXX were visualized as XXXXX and pathways', ['ppi information', 'interaction networks'])\n", "('NONE', 'we characterized the pathway-level XXXXX of XXXXX', ['topological properties', 'drug targets'])\n", "('method used for task', 'we optimized XXXXX based on pathway-level XXXXX', ['drug targets', 'topological properties'])\n", "('method used for task', 'XXXXX and XXXXX perform best in terms of auc', ['random forests', 'deep neural networks'])\n", "('NONE', 'XXXXX helps identify patients’ XXXXX and issues', ['distributed cognition', 'interaction strategies'])\n", "('method used for task', 'we proposed a XXXXX to automatically de-identify XXXXX', ['hybrid system', 'electronic medical records'])\n", "('method used for task', 'supervised XXXXX to identify XXXXX for heart disease in ehrs', ['information extraction', 'risk factors'])\n", "('method used for task', 'supervised XXXXX to identify risk factors for XXXXX in ehrs', ['information extraction', 'heart disease'])\n", "('method used for task', 'supervised information extraction to identify XXXXX for XXXXX in ehrs', ['risk factors', 'heart disease'])\n", "('method used for task', 'XXXXX and glass box evaluations are both needed to develop XXXXX', ['black box', 'complex systems'])\n", "('method used for task', 'XXXXX and XXXXX evaluations are both needed to develop complex systems', ['black box', 'glass box'])\n", "('method used for task', 'black box and XXXXX evaluations are both needed to develop XXXXX', ['complex systems', 'glass box'])\n", "('NONE', 'XXXXX of XXXXX using approximate randomization techniques', ['statistical testing', 'model performance'])\n", "('method used for task', 'we used XXXXX ( nlp ) to extract XXXXX', ['natural language processing', 'heart disease risk factors'])\n", "('method used for task', 'developed a rule-based text mining system to identify framingham XXXXX required for predicting XXXXX', ['risk factors', 'coronary artery disease'])\n", "('method used for task', 'developed a rule-based XXXXX to identify framingham XXXXX required for predicting coronary artery disease', ['risk factors', 'text mining system'])\n", "('method used for task', 'developed a rule-based XXXXX to identify framingham risk factors required for predicting XXXXX', ['coronary artery disease', 'text mining system'])\n", "('method used for task', 'XXXXX to identify XXXXX in diabetic patients over time', ['nlp system', 'heart disease risk factors'])\n", "('NONE', 'XXXXX in XXXXX ( emr ) was automated', ['electronic medical records', 'risk factor detection'])\n", "('NONE', 'modeling context with XXXXX yielded better XXXXX', ['distributional semantics', 'predictive models'])\n", "('method used for task', 'we proposed a XXXXX to automatically identify XXXXX', ['hybrid system', 'heart disease risk factors'])\n", "('NONE', 'XXXXX is of great importance for the treatment of XXXXX', ['heart disease', 'risk factor detection'])\n", "('NONE', 'the experimental set up used 6 XXXXX with 7 different XXXXX', ['machine learning models', 'feature sets'])\n", "('NONE', 'fall-injury XXXXX systems are static and provide no form of XXXXX', ['user interaction', 'prevention intervention'])\n", "('method used for task', 'cross-fall XXXXX systems face similar challenges to other fall XXXXX', ['prevention intervention', 'prevention systems'])\n", "('method used for task', 'cross-fall XXXXX systems face similar challenges to other fall XXXXX', ['prevention intervention', 'prevention systems'])\n", "('method used for task', 'the XXXXX can be modified to accommodate XXXXX', ['classification system', 'technological development'])\n", "('method used for task', 'a XXXXX and a hybrid type ii XXXXX are applied', ['fuzzy system', 'data mining technique'])\n", "('method used for task', 'two different types of XXXXX were created to generate a XXXXX', ['membership functions', 'hybrid system'])\n", "('method used for task', 'a XXXXX is established to describe the XXXXX', ['mathematical model', 'localization problem'])\n", "('NONE', 'introduction of XXXXX as a bridge between measurement input and analytical XXXXX for manufacturing', ['repair features', 'cad data'])\n", "('method used for task', 'dynamic determination and adjustment of XXXXX and cax repair process chain execution based on XXXXX', ['function blocks', 'repair features'])\n", "('method used for task', 'dynamic determination and adjustment of repair features and cax XXXXX chain execution based on XXXXX', ['function blocks', 'repair process'])\n", "('method used for task', 'dynamic determination and adjustment of XXXXX and cax XXXXX chain execution based on function blocks', ['repair features', 'repair process'])\n", "('NONE', 'XXXXX study of repairing worn-out turbine blades with repair features and XXXXX', ['use case', 'function blocks'])\n", "('NONE', 'XXXXX study of repairing worn-out turbine blades with XXXXX and function blocks', ['use case', 'repair features'])\n", "('method used for task', 'use case study of repairing worn-out turbine blades with XXXXX and XXXXX', ['function blocks', 'repair features'])\n", "('method used for task', 'in this paper , an XXXXX is proposed for solving minimum time manufacturing XXXXX in multi points manufacturing tasks', ['path planning', 'optimization strategy'])\n", "('method used for task', 'according to the start-stop movement in drilling/spot welding task , the XXXXX problem is converted into a XXXXX ( tsp ) and a series of point to point minimum time transfer XXXXX problems', ['traveling salesman problem', 'path planning'])\n", "('method used for task', 'according to the start-stop movement in drilling/spot welding task , the XXXXX is converted into a XXXXX ( tsp ) and a series of point to point minimum time transfer path planning problems', ['traveling salesman problem', 'path planning problem'])\n", "('method used for task', 'according to the start-stop movement in drilling/spot welding task , the XXXXX problem is converted into a traveling salesman problem ( tsp ) and a series of point to point minimum time transfer XXXXX problems', ['path planning', 'path planning problem'])\n", "('NONE', 'the XXXXX are automatically generated from laser-scanned XXXXX', ['point clouds', 'environment models'])\n", "('NONE', 'performed XXXXX by XXXXX and experimental of live industrial problem', ['design optimization', 'finite element analysis'])\n", "('method used for task', 'this paper enables a “step by step” process to identify the most appropriate XXXXX for a company’s pss problem . an example is also introduced to illustrate the use of the proposed scoring criteria and provide a clear picture of how different XXXXX can be utilized at their best in terms of application', ['design methodology', 'design methodologies'])\n", "('method used for task', 'this paper enables a “step by step” process to identify the most appropriate XXXXX for a company’s pss problem . an example is also introduced to illustrate the use of the proposed XXXXX and provide a clear picture of how different design methodologies can be utilized at their best in terms of application', ['design methodology', 'scoring criteria'])\n", "('NONE', 'this paper enables a “step by step” process to identify the most appropriate design methodology for a company’s pss problem . an example is also introduced to illustrate the use of the proposed XXXXX and provide a clear picture of how different XXXXX can be utilized at their best in terms of application', ['design methodologies', 'scoring criteria'])\n", "('NONE', 'an example is also introduced to illustrate the use of the proposed XXXXX and provide a clear picture of how different XXXXX can be utilized at their best in terms of application', ['design methodologies', 'scoring criteria'])\n", "('method used for task', 'XXXXX and XXXXX scheduling are major issues in optimization of hole-making operations', ['tool travel', 'tool switch'])\n", "('method used for task', 'it is necessary to achieve a correct sequence of hole-making operations to minimize XXXXX and XXXXX and finally to minimize total processing cost', ['tool travel', 'tool switch'])\n", "('NONE', \"we present an efficient technique for sketch-based XXXXX using automatically extracted XXXXX . an automatic and real-time method is proposed to align a user 's hand-drawn sketch line to the contour lines of an image , facilitating a considerable level of ease for XXXXX\", ['3d modeling', 'image features'])\n", "('method used for task', \"we present an efficient technique for sketch-based XXXXX using automatically extracted image features . an automatic and real-time method is proposed to align a user 's hand-drawn sketch line to the XXXXX of an image , facilitating a considerable level of ease for XXXXX\", ['3d modeling', 'contour lines'])\n", "('method used for task', \"we present an XXXXX for sketch-based XXXXX using automatically extracted image features . an automatic and real-time method is proposed to align a user 's hand-drawn sketch line to the contour lines of an image , facilitating a considerable level of ease for XXXXX\", ['3d modeling', 'efficient technique'])\n", "('method used for task', \"we present an efficient technique for sketch-based XXXXX using automatically extracted image features . an automatic and real-time method is proposed to align a user 's hand-drawn XXXXX to the contour lines of an image , facilitating a considerable level of ease for XXXXX\", ['3d modeling', 'sketch line'])\n", "('method used for task', \"we present an efficient technique for sketch-based 3d modeling using automatically extracted XXXXX . an automatic and real-time method is proposed to align a user 's hand-drawn sketch line to the XXXXX of an image , facilitating a considerable level of ease for 3d modeling\", ['image features', 'contour lines'])\n", "('NONE', \"we present an efficient technique for sketch-based XXXXX using automatically extracted XXXXX . an automatic and real-time method is proposed to align a user 's hand-drawn sketch line to the contour lines of an image , facilitating a considerable level of ease for XXXXX\", ['image features', '3d modeling'])\n", "('method used for task', \"we present an XXXXX for sketch-based 3d modeling using automatically extracted XXXXX . an automatic and real-time method is proposed to align a user 's hand-drawn sketch line to the contour lines of an image , facilitating a considerable level of ease for 3d modeling\", ['image features', 'efficient technique'])\n", "('method used for task', \"we present an efficient technique for sketch-based 3d modeling using automatically extracted XXXXX . an automatic and real-time method is proposed to align a user 's hand-drawn XXXXX to the contour lines of an image , facilitating a considerable level of ease for 3d modeling\", ['image features', 'sketch line'])\n", "('method used for task', \"we present an efficient technique for sketch-based XXXXX using automatically extracted image features . an automatic and real-time method is proposed to align a user 's hand-drawn sketch line to the XXXXX of an image , facilitating a considerable level of ease for XXXXX\", ['contour lines', '3d modeling'])\n", "('method used for task', \"we present an XXXXX for sketch-based 3d modeling using automatically extracted image features . an automatic and real-time method is proposed to align a user 's hand-drawn sketch line to the XXXXX of an image , facilitating a considerable level of ease for 3d modeling\", ['contour lines', 'efficient technique'])\n", "('method used for task', \"we present an efficient technique for sketch-based 3d modeling using automatically extracted image features . an automatic and real-time method is proposed to align a user 's hand-drawn XXXXX to the XXXXX of an image , facilitating a considerable level of ease for 3d modeling\", ['contour lines', 'sketch line'])\n", "('method used for task', \"we present an XXXXX for sketch-based XXXXX using automatically extracted image features . an automatic and real-time method is proposed to align a user 's hand-drawn sketch line to the contour lines of an image , facilitating a considerable level of ease for XXXXX\", ['3d modeling', 'efficient technique'])\n", "('method used for task', \"we present an efficient technique for sketch-based XXXXX using automatically extracted image features . an automatic and real-time method is proposed to align a user 's hand-drawn XXXXX to the contour lines of an image , facilitating a considerable level of ease for XXXXX\", ['3d modeling', 'sketch line'])\n", "('method used for task', \"we present an XXXXX for sketch-based 3d modeling using automatically extracted image features . an automatic and real-time method is proposed to align a user 's hand-drawn XXXXX to the contour lines of an image , facilitating a considerable level of ease for 3d modeling\", ['efficient technique', 'sketch line'])\n", "('NONE', 'we use a XXXXX to align a sketch line to the outlines of an image using the features of the sketch line and XXXXX of an image , and some operations are proposed to refine the result of alignment', ['geometric method', 'contour lines'])\n", "('method used for task', 'we use a XXXXX to align a XXXXX to the outlines of an image using the features of the XXXXX and contour lines of an image , and some operations are proposed to refine the result of alignment', ['geometric method', 'sketch line'])\n", "('method used for task', 'we use a XXXXX to align a XXXXX to the outlines of an image using the features of the XXXXX and contour lines of an image , and some operations are proposed to refine the result of alignment', ['geometric method', 'sketch line'])\n", "('NONE', 'we use a geometric method to align a XXXXX to the outlines of an image using the features of the XXXXX and XXXXX of an image , and some operations are proposed to refine the result of alignment', ['contour lines', 'sketch line'])\n", "('NONE', 'we use a geometric method to align a XXXXX to the outlines of an image using the features of the XXXXX and XXXXX of an image , and some operations are proposed to refine the result of alignment', ['contour lines', 'sketch line'])\n", "('NONE', 'in the sketch-based XXXXX , the XXXXX is represented by a editable spline , therefore , the aligned XXXXX can be further adjusted interactively', ['3d modeling method', 'sketch line'])\n", "('NONE', 'in the sketch-based XXXXX , the XXXXX is represented by a editable spline , therefore , the aligned XXXXX can be further adjusted interactively', ['3d modeling method', 'sketch line'])\n", "('NONE', 'XXXXX of the XXXXX on the surface of a sphere', ['numerical solution', 'monge–ampère equation'])\n", "('method used for task', 'the XXXXX applies simultaneously to fluid mechanics and XXXXX', ['mathematical model', 'solid mechanics'])\n", "('method used for task', 'we present an automated ensemble XXXXX for modelling cerebrovascular XXXXX under a range of exercise intensities', ['blood flow', 'simulation method'])\n", "('method used for task', 'software for modeling uncertainty using XXXXX and XXXXX expansions', ['monte carlo', 'polynomial chaos'])\n", "('NONE', 'increasing XXXXX will enhance XXXXX', ['confinement ratio', 'droplet deformation'])\n", "('NONE', 'the droplet is found to orient more towards the flow direction with increasing XXXXX or XXXXX', ['viscosity ratio', 'confinement ratio'])\n", "('method used for task', 'the transparency and robustness are considered as an XXXXX and solved by applying XXXXX', ['optimization problem', 'genetic algorithms'])\n", "('method used for task', 'a XXXXX for 3-d XXXXX is presented', ['mathematical model', 'computer graphs'])\n", "('method used for task', 'a theoretical approach of a XXXXX for 3-d XXXXX and corresponding methods for empirical testing are presented', ['mathematical model', 'computer graphs'])\n", "('method used for task', 'proposing an approach that can provide admissions control for all vod systems . it addresses the following challenges : XXXXX and several algorithms for XXXXX', ['resource allocation', 'scheduling policies'])\n", "('NONE', 'autogrow is an XXXXX that facilitates XXXXX and optimization', ['evolutionary algorithm', 'drug design'])\n", "('NONE', 'the XXXXX leverages strengths of two XXXXX , rtc and railsys', ['hybrid approach', 'simulation tools'])\n", "('method used for task', 'XXXXX to assess the frequency and size of XXXXX', ['empirical analysis', 'coherent clusters'])\n", "('method used for task', 'a study on the relationship between XXXXX and XXXXX', ['coherent clusters', 'system evolution'])\n", "('NONE', 'the approach was assessed in four XXXXX in two parallel XXXXX', ['case studies', 'research projects'])\n", "('method used for task', 'evaluation of the approach includes XXXXX and XXXXX', ['comparative analysis', 'usability studies'])\n", "('method used for task', 'XXXXX is knowledge-intensive and XXXXX is crucial', ['software engineering', 'intellectual capital'])\n", "('method used for task', 'the XXXXX is increased while the XXXXX is reduced', ['classification rate', 'texture feature space'])\n", "('NONE', 'XXXXX based on assessment of XXXXX through domain expert interviews', ['system design', 'user needs'])\n", "('NONE', 'XXXXX via smartphone facilitate the XXXXX', ['collaborative work', 'emergency management'])\n", "('method used for task', 'XXXXX for XXXXX and visual reasoning using taxonomy', ['conceptual graphs', 'knowledge representation'])\n", "('NONE', 'XXXXX improves the XXXXX of the credit ratings', ['classification accuracy', 'feature selection procedure'])\n", "('NONE', 'a XXXXX can enhance the XXXXX between particles', ['crossover operator', 'information exchange'])\n", "('method used for task', 'an efficient XXXXX based without center set according to XXXXX of the network', ['structural properties', 'community detection algorithm'])\n", "('method used for task', 'a simple XXXXX is proposed to construct a XXXXX', ['two-stage approach', 'fuzzy regression model'])\n", "('NONE', 'the XXXXX contains the XXXXX and a fuzzy adjustment term', ['fuzzy model', 'crisp coefficients'])\n", "('method used for task', 'we propose a hybrid artificial bee XXXXX for the cyclic XXXXX', ['colony algorithm', 'antibandwidth problem'])\n", "('NONE', 'the XXXXX can get a XXXXX for small regularization values', ['decomposition method', 'fast convergence'])\n", "('method used for task', 'constructing a XXXXX and XXXXX with heterogeneous items firstly', ['joint replenishment', 'delivery model'])\n", "('NONE', 'a novel XXXXX metric called XXXXX ( gig ) is proposed', ['feature selection', 'global information gain'])\n", "('NONE', 'an XXXXX called maximizing XXXXX ( mgig ) is developed', ['efficient algorithm', 'global information gain'])\n", "('NONE', 'a two-phase cost-sensitive XXXXX combining with XXXXX', ['emotion classification', 'topic detection'])\n", "('method used for task', 'we develop a XXXXX agr-sce to find the XXXXX', ['heuristic algorithm', 'generalization reduct'])\n", "('NONE', 'the XXXXX was utilized to find similar users in the XXXXX', ['social network', 'collaborative filtering'])\n", "('NONE', 'we first analyze the shortages of the existing XXXXX in XXXXX', ['similarity measures', 'collaborative filtering'])\n", "('NONE', 'we demonstrated the occurrence of XXXXX in different XXXXX in dea', ['rank reversal', 'ranking models'])\n", "('NONE', 'our method outperforms state-of-the-art solutions on XXXXX , achieving in particular a XXXXX', ['high precision', 'benchmark data'])\n", "('method used for task', 'it uses a XXXXX to provide personalized treatments to patients with XXXXX and any advices to prevent it', ['recommender system', 'low back pain'])\n", "('method used for task', 'XXXXX is regarded as a multi-index XXXXX', ['feature selection', 'evaluation process'])\n", "('method used for task', 'we apply a two-stage XXXXX to evaluate XXXXX and revenue efficiency', ['cost efficiency', 'dea model'])\n", "('method used for task', 'we develop a XXXXX to deal with uncertain or vague XXXXX', ['fuzzy ontology', 'knowledge representation'])\n", "('NONE', 'the integration of the dual XXXXX gifts the cso algorithm with powerful XXXXX', ['search mechanisms', 'global search ability'])\n", "('method used for task', 'new XXXXX models are based on full XXXXX', ['probability density functions', 'key profile'])\n", "('NONE', 'we construct a margin related XXXXX to learn the weights of XXXXX', ['loss function', 'base classifiers'])\n", "('NONE', 'the system was improved with XXXXX , XXXXX and statistical techniques', ['text mining', 'neural networks'])\n", "('method used for task', 'the system was improved with XXXXX , neural networks and XXXXX', ['text mining', 'statistical techniques'])\n", "('method used for task', 'the system was improved with text mining , XXXXX and XXXXX', ['neural networks', 'statistical techniques'])\n", "('NONE', 'new design for reliable XXXXX with higher XXXXX and quality', ['inference system', 'software reliability'])\n", "('method used for task', 'introduces a novel XXXXX based on bagged and XXXXX', ['fuzzy clustering', 'segmentation method'])\n", "('method used for task', 'proposes how to adapt bagged XXXXX for XXXXX', ['fuzzy clustering', 'fuzzy data'])\n", "('NONE', 'a fusion of iso-clahe improved the XXXXX from a low lighting or XXXXX', ['contrast ratio', 'eye image'])\n", "('method used for task', 'an innovative XXXXX is proposed for diagnosing XXXXX', ['intelligent system', 'heart diseases'])\n", "('method used for task', 'ctc combined with smote tops state of the XXXXX designed to tackle XXXXX', ['class imbalance', 'art techniques'])\n", "('NONE', 'an XXXXX of XXXXX and classification methods is developed', ['experimental study', 'feature extraction'])\n", "('NONE', 'an XXXXX of feature extraction and XXXXX is developed', ['experimental study', 'classification methods'])\n", "('method used for task', 'an experimental study of XXXXX and XXXXX is developed', ['feature extraction', 'classification methods'])\n", "('method used for task', 'XXXXX conclude that isso outperforms abc for 50 XXXXX', ['numerical examples', 'benchmark functions'])\n", "('NONE', 'novel sudden XXXXX index ( scdi ) is proposed using XXXXX', ['ecg signals', 'cardiac death'])\n", "('method used for task', 'a multi-attribute XXXXX is proposed by combining XXXXX and topsis', ['relative entropy', 'ranking method'])\n", "('method used for task', 'XXXXX and XXXXX based mil algorithm is proposed', ['extreme learning machine', 'classifier ensemble'])\n", "('NONE', 'a new method of XXXXX of XXXXX is presented', ['automatic verification', 'knowledge base'])\n", "('method used for task', 'verification looks for discrepancy between the XXXXX and XXXXX', ['expert knowledge', 'knowledge base'])\n", "('NONE', 'experts and XXXXX system use the half-marks in XXXXX', ['automatic verification', 'inference rules'])\n", "('method used for task', 'we propose a fast XXXXX to solve the modeled XXXXX', ['memetic algorithm', 'optimization problems'])\n", "('method used for task', 'a XXXXX is used to show the applicability of the proposed XXXXX', ['case study', 'optimization model'])\n", "('method used for task', 'XXXXX as regards key parameters is performed in the XXXXX', ['sensitivity analysis', 'case study'])\n", "('method used for task', 'the approach is based on XXXXX and XXXXX', ['semantic technologies', 'web standards'])\n", "('NONE', 'a XXXXX describes experiences with a XXXXX in industrial use for years', ['case study', 'decision support system'])\n", "('NONE', 'XXXXX and classification of XXXXX using hadoop framework', ['feature selection', 'microarray data'])\n", "('method used for task', 'mapreduce based XXXXX are proposed for XXXXX ( fs )', ['statistical tests', 'feature selection'])\n", "('method used for task', 'XXXXX are converted to 1d signals using XXXXX', ['fundus images', 'radon transform'])\n", "('method used for task', 'the novel node coupling XXXXX for XXXXX are proposed', ['clustering methods', 'link prediction'])\n", "('method used for task', 'we exploit XXXXX shapelets for complex XXXXX', ['time series', 'human activity recognition'])\n", "('method used for task', 'we implement a XXXXX based on smartphone for XXXXX', ['human activity recognition', 'prototype system'])\n", "('NONE', 'ldmdba has a XXXXX of ( logdlogn ) for predicting a new XXXXX', ['time complexity', 'data point'])\n", "('NONE', 'we apply XXXXX in XXXXX to make the best use of historical driving data', ['data mining algorithms', 'train operations'])\n", "('NONE', 'two sto algorithms are proposed by combining XXXXX , XXXXX and train parking methods', ['expert knowledge', 'data mining'])\n", "('method used for task', 'XXXXX and XXXXX are conducted to present the findings', ['numerical simulation', 'empirical studies'])\n", "('NONE', 'a new total XXXXX in XXXXX is proposed', ['uncertainty measure', 'evidence theory'])\n", "('method used for task', 'the rimer-based XXXXX is fine-tuned and validated using XXXXX', ['prediction model', 'historical data'])\n", "('method used for task', 'a novel XXXXX is used to enhance the XXXXX', ['learning strategy', 'global search ability'])\n", "('NONE', 'it proposes a XXXXX qos dynamic XXXXX based on improvedâ case-based reasoning ( cbr )', ['web service', 'prediction method'])\n", "('method used for task', 'it proposes a XXXXX qos dynamic prediction method based on improvedâ XXXXX ( cbr )', ['web service', 'case-based reasoning'])\n", "('method used for task', 'it proposes a web service qos dynamic XXXXX based on improvedâ XXXXX ( cbr )', ['prediction method', 'case-based reasoning'])\n", "('method used for task', 'an opposite-based approach to XXXXX is presented as a unifying view to basic XXXXX', ['fuzzy modeling', 'knowledge representation'])\n", "('method used for task', 'we claim that paired concepts are the basic XXXXX for XXXXX', ['building blocks', 'knowledge representation'])\n", "('NONE', 'we extract a XXXXX from event logs using XXXXX', ['process mining', 'process structure'])\n", "('NONE', 'XXXXX improves the efficiency in XXXXX', ['sample selection', 'feature dimension reduction'])\n", "('NONE', 'bayesian XXXXX of a XXXXX using a novel mcmc algorithm', ['system identification', 'nonlinear dynamical system'])\n", "('NONE', 'data annealing is applied to the XXXXX of a XXXXX', ['system identification', 'nonlinear dynamical system'])\n", "('NONE', 'techniques are developed which allow one to choose a subset of the available training data which is ‘highly informative’ with regards to both levels of XXXXX – XXXXX and model selection', ['bayesian inference', 'parameter estimation'])\n", "('method used for task', 'techniques are developed which allow one to choose a subset of the available training data which is ‘highly informative’ with regards to both levels of XXXXX – parameter estimation and XXXXX', ['bayesian inference', 'model selection'])\n", "('method used for task', 'techniques are developed which allow one to choose a subset of the available training data which is ‘highly informative’ with regards to both levels of bayesian inference – XXXXX and XXXXX', ['parameter estimation', 'model selection'])\n", "('NONE', 'virtual and XXXXX of the controllers on a fully XXXXX', ['experimental validation', 'electric vehicle'])\n", "('NONE', 'a XXXXX of myocardial XXXXX imaging is proposed', ['finite element model', 'mr perfusion'])\n", "('method used for task', 'XXXXX are evaluated directly from the 4d XXXXX , without the need of any computational mesh', ['pressure differences', 'flow data'])\n", "('NONE', 'we present a standardisation for evaluating XXXXX for late enhancement imaging of infarct in the XXXXX', ['left ventricle', 'segmentation algorithms'])\n", "('method used for task', 'we provide a consensus ground truth obtained with XXXXX on the XXXXX . future algorithms can thus be benchmarked to provide a more reliable result', ['statistical modelling', 'benchmark datasets'])\n", "('NONE', 'the XXXXX has a positive correlation with the XXXXX', ['tensile strength', 'strain rate'])\n", "('NONE', 'XXXXX in the bulk solder plays dominant role at the low XXXXX', ['ductile fracture', 'strain rate'])\n", "('NONE', 'an XXXXX on bond wire looping by using high-speed XXXXX was presented', ['experimental study', 'video analysis'])\n", "('NONE', 'compare scanning acoustic XXXXX with the XXXXX throughout a power cycling test which help to build correlation between the two methods', ['structure function', 'microscopy images'])\n", "('method used for task', 'XXXXX and XXXXX are applied for improvement', ['arq dynamics', 'crc codes'])\n", "('method used for task', 'XXXXX and XXXXX are applied for improvement', ['arq dynamics', 'crc codes'])\n", "('method used for task', 'an efficient XXXXX for a dendrite morphological XXXXX', ['training algorithm', 'neural network'])\n", "('NONE', 'the proposed algorithm is much less XXXXX than the trace norm minimization algorithm especially facing the XXXXX', ['computational cost', 'large data'])\n", "('NONE', 'we proposed a XXXXX by using a selective XXXXX technique to reduce dsp complexity', ['mimo system', 'mode excitation'])\n", "('method used for task', 'a domain specific language and XXXXX for on- and off-line XXXXX', ['data analytics', 'runtime models'])\n", "('NONE', 'our method maximises the XXXXX between a small set of XXXXX', ['euclidean distance', 'feature points'])\n", "('method used for task', 'ensemble XXXXX based on XXXXX of texture features was effective', ['svm classification', 'sparse coding'])\n", "('NONE', 'ensemble XXXXX based on sparse coding of XXXXX was effective', ['svm classification', 'texture features'])\n", "('NONE', 'ensemble svm classification based on XXXXX of XXXXX was effective', ['sparse coding', 'texture features'])\n", "('NONE', 'we utilise a novel social connections metric and a XXXXX as XXXXX', ['contextual information', 'scene model'])\n", "('method used for task', 'a XXXXX tool for robotized XXXXX has been developed and constructed', ['cable feeder', 'cable winding'])\n", "('method used for task', 'development of a framework for XXXXX and in-service data feedback to XXXXX', ['knowledge management', 'product development'])\n", "('method used for task', 'demonstrating how XXXXX is captured , fed back and reused for XXXXX', ['field data', 'product development'])\n", "('NONE', 'the paper provides XXXXX of path trajectory generation and ndt XXXXX on two large , curved surfaces of a composite aerofoil component', ['experimental validation', 'data acquisition'])\n", "('NONE', 'the paper provides XXXXX of path trajectory generation and ndt data acquisition on two large , XXXXX of a composite aerofoil component', ['experimental validation', 'curved surfaces'])\n", "('NONE', 'the paper provides experimental validation of path trajectory generation and ndt XXXXX on two large , XXXXX of a composite aerofoil component', ['data acquisition', 'curved surfaces'])\n", "('method used for task', 'XXXXX for multi-level XXXXX', ['hierarchical model', 'adaptive systems'])\n", "('NONE', 'application to the XXXXX of atvs XXXXX', ['case study', 'motion control'])\n", "('method used for task', 'XXXXX is robust to XXXXX and video frame-based attacks', ['signal processing', 'watermark scheme'])\n", "('NONE', 'XXXXX of costas loop in the signal׳s XXXXX is demonstrated', ['nonlinear analysis', 'phase space'])\n", "('NONE', 'an adaptive multi-view feature selection ( amfs ) algorithm is proposed to fuse XXXXX formotion XXXXX', ['multiple features', 'data retrieval'])\n", "('method used for task', 'XXXXX for the binary m-qim ( multi-symbol XXXXX modulation )', ['theoretical framework', 'quantization index'])\n", "('method used for task', 'the difficulty in XXXXX is to balance the simulation cost and XXXXX', ['spot market', 'execution time'])\n", "('NONE', 'an analysis of key drivers governing XXXXX on XXXXX has been carried out', ['amazon ec2', 'spot prices'])\n", "('method used for task', 'we give a survey of 24 XXXXX for text-dependent XXXXX', ['speaker verification', 'speech databases'])\n", "('method used for task', 'we present a method for interpolation between XXXXX for XXXXX', ['speech synthesis', 'language varieties'])\n", "('method used for task', 'a layered algorithm integrates XXXXX and XXXXX', ['dynamic bayesian network', 'supervised clustering'])\n", "('method used for task', 'XXXXX shows the great value to XXXXX in data mining', ['supervised clustering', 'feature reduction'])\n", "('method used for task', 'XXXXX shows the great value to feature reduction in XXXXX', ['supervised clustering', 'data mining'])\n", "('NONE', 'supervised clustering shows the great value to XXXXX in XXXXX', ['feature reduction', 'data mining'])\n", "('method used for task', 'XXXXX for transport mode detection studies from XXXXX using coefficient of variation cv', ['sample size', 'gps data'])\n", "('NONE', 'the proposed XXXXX ( ma ) is a combination of a XXXXX and a simulated annealing approach', ['memetic algorithm', 'genetic algorithm'])\n", "('method used for task', 'a XXXXX for evaluating XXXXX locations is proposed', ['generic model', 'traffic accident'])\n" ] }, { "data": { "text/plain": [ "array(['NONE', 'method used for task', 'method used for task', ...,\n", " 'method used for task', 'NONE', 'method used for task'],\n", " dtype=', 'encoding': 'utf-8', 'input': 'content', 'lowercase': True, 'max_df': 1.0, 'max_features': None, 'min_df': 1, 'ngram_range': (1, 1), 'preprocessor': None, 'stop_words': None, 'strip_accents': None, 'token_pattern': '(?u)\\\\b\\\\w\\\\w+\\\\b', 'tokenizer': None, 'vocabulary': None}\n", "\t-0.9520\tof \t\t1.0542\tis \n", "\t-0.4352\tspecified \t\t0.9787\tto \n", "\t-0.4352\tusing \t\t0.8733\tfor \n", "\t-0.4313\tann \t\t0.4851\tand \n", "\t-0.4313\tfind \t\t0.4785\tsolved \n", "\t-0.3274\tdecreases \t\t0.4785\tassists \n", "\t-0.3258\tthat \t\t0.4309\tare \n", "\t-0.3181\tallowing \t\t0.4151\tsolve \n", "\t-0.3181\texcept \t\t0.4081\ton \n", "\t-0.3074\tas \t\t0.4081\tapplication \n", "\t-0.2935\tin \t\t0.3915\tmore \n", "\t-0.2892\tintroduced \t\t0.3915\tcapable \n", "\t-0.2892\tunified \t\t0.3526\tpresented \n", "\t-0.2755\tchecks \t\t0.3343\tadopted \n", "\t-0.2755\tcollision \t\t0.3208\tbuilding \n", "\t-0.2755\toverall \t\t0.3208\t3d \n", "\t-0.2755\tvirtual \t\t0.2736\tga \n", "\t-0.2715\tcan \t\t0.2736\tautomated \n", "\t-0.2715\tensure \t\t0.2398\tsizing \n", "\t-0.2651\talgorithms \t\t0.2240\tpso \n", "('NONE', 'a method for estimation of XXXXX of XXXXX is presented', ['effective properties', 'porous materials'])\n", "('method used for task', 'accounting for XXXXX is essential for estimation of XXXXX', ['nonlinear effects', 'effective properties'])\n", "('method used for task', 'develops the heterogeneous XXXXX for fiber-reinforced XXXXX', ['feature model', 'object modeling'])\n", "('NONE', 'two formulations for the problem of optimum XXXXX of onshore XXXXX', ['layout design', 'wind farms'])\n", "('method used for task', 'boundary-value and initial-value XXXXX are solved using XXXXX and graph products', ['differential equations', 'finite difference method'])\n", "('method used for task', 'boundary-value and initial-value XXXXX are solved using finite difference method and XXXXX', ['differential equations', 'graph products'])\n", "('method used for task', 'boundary-value and initial-value differential equations are solved using XXXXX and XXXXX', ['finite difference method', 'graph products'])\n", "('method used for task', 'an XXXXX to couple cfd and XXXXX is presented', ['open-source software', 'multiobjective optimization'])\n", "('method used for task', 'parallel XXXXX for solving cfd XXXXX is implemented', ['evolutionary algorithm', 'optimization problems'])\n", "('method used for task', 'a novel XXXXX platform is developed for the examination of an rfid-enabled mascs in a flexible XXXXX , and several system performance measures are considered in the XXXXX platform', ['assembly line', 'simulation test'])\n", "('method used for task', 'a novel simulation test platform is developed for the examination of an rfid-enabled mascs in a flexible XXXXX , and several XXXXX are considered in the simulation test platform', ['assembly line', 'system performance measures'])\n", "('method used for task', 'a novel XXXXX platform is developed for the examination of an rfid-enabled mascs in a flexible XXXXX , and several system performance measures are considered in the XXXXX platform', ['assembly line', 'simulation test'])\n", "('method used for task', 'a novel XXXXX platform is developed for the examination of an rfid-enabled mascs in a flexible assembly line , and several system performance measures are considered in the XXXXX platform', ['simulation test', 'simulation platform'])\n", "('method used for task', 'a novel XXXXX platform is developed for the examination of an rfid-enabled mascs in a flexible assembly line , and several XXXXX are considered in the XXXXX platform', ['simulation test', 'system performance measures'])\n", "('method used for task', 'a novel XXXXX platform is developed for the examination of an rfid-enabled mascs in a flexible assembly line , and several system performance measures are considered in the XXXXX platform', ['simulation test', 'simulation platform'])\n", "('method used for task', 'a novel XXXXX platform is developed for the examination of an rfid-enabled mascs in a flexible assembly line , and several system performance measures are considered in the XXXXX platform', ['simulation platform', 'simulation test'])\n", "('method used for task', 'a novel XXXXX platform is developed for the examination of an rfid-enabled mascs in a flexible assembly line , and several XXXXX are considered in the XXXXX platform', ['system performance measures', 'simulation test'])\n", "('method used for task', 'a novel XXXXX platform is developed for the examination of an rfid-enabled mascs in a flexible assembly line , and several system performance measures are considered in the XXXXX platform', ['simulation test', 'simulation platform'])\n", "('NONE', 'the XXXXX provides more efficient XXXXX', ['information exchange', 'reduced-order model server'])\n", "('method used for task', 'XXXXX and XXXXX that make the approach practical are presented', ['software design', 'implementation issues'])\n", "('NONE', 'the XXXXX of XXXXX can be generated , which can be used for the shape design of electrodes', ['cad model', 'volume feature'])\n", "('method used for task', 'the related parameters of XXXXX are extracted , which can be used for the XXXXX of electrodes and the setting of processing parameters', ['parametric design', 'volume feature'])\n", "('method used for task', 'an XXXXX is proposed for XXXXX of steel space frames under lrfd-aisc provisions', ['efficient algorithm', 'optimum design'])\n", "('method used for task', 'geometric nonlinear behavior of XXXXX is considered during the XXXXX', ['steel frames', 'optimization process'])\n", "('NONE', 'icde is extended to the XXXXX of XXXXX by combining icde with sora which gives a so-called the sora-icde', ['truss structures', 'rbdo problem'])\n", "('NONE', 'numerical results for five XXXXX illustrate the effectiveness of the sora-icde in solving the rbdo problem of XXXXX', ['benchmark problems', 'truss structures'])\n", "('NONE', 'numerical results for five XXXXX illustrate the effectiveness of the sora-icde in solving the XXXXX of truss structures', ['benchmark problems', 'rbdo problem'])\n", "('NONE', 'numerical results for five benchmark problems illustrate the effectiveness of the sora-icde in solving the XXXXX of XXXXX', ['truss structures', 'rbdo problem'])\n", "('NONE', 'XXXXX of the wheel is estimated using stress life ( s-n ) method based on the XXXXX of the wheel', ['fatigue life', 'stress analysis'])\n", "('NONE', 'simulations of ballistic XXXXX were performed using the validated XXXXX', ['impact tests', 'fe model'])\n", "('NONE', 'static and XXXXX of variable geometry microbeams using XXXXX', ['dynamic analysis', 'energy method'])\n", "('method used for task', 'a XXXXX is proposed to handle thin-walled model with XXXXX', ['hybrid method', 'complex geometry'])\n", "('method used for task', 'the XXXXX of woa is confirmed by the results on XXXXX', ['multimodal functions', 'exploration ability'])\n", "('method used for task', 'this paper develops large scale marine hydrological environmental data-oriented XXXXX and realizes oceanographic planar graph , contour line rendering , isosurface rendering , factor field volume rendering and XXXXX of current field', ['dynamic simulation', 'visualization software'])\n", "('NONE', 'the system employs cuda XXXXX to improve the computation rate of XXXXX of marine water environmental factors based on netcdf ( network common data form ) format', ['parallel computing', 'volume rendering'])\n", "('method used for task', 'the system employs cuda XXXXX to improve the XXXXX of volume rendering of marine water environmental factors based on netcdf ( network common data form ) format', ['parallel computing', 'computation rate'])\n", "('NONE', 'the system employs cuda parallel computing to improve the XXXXX of XXXXX of marine water environmental factors based on netcdf ( network common data form ) format', ['volume rendering', 'computation rate'])\n", "('method used for task', 'nuclear XXXXX modeling generality and robustness are improved by a modular , agent based XXXXX', ['modeling framework', 'fuel cycle'])\n", "('NONE', 'paper reviews applications of XXXXX in XXXXX', ['artificial neural networks', 'model calibration'])\n", "('method used for task', 'worst XXXXX oriented approach for volume fraction minimization in XXXXX with uncertain load directions and compliance constraints is considered', ['topology optimization', 'load direction'])\n", "('method used for task', 'the stored data from XXXXX are replaced by continuous functions suitable for representation in XXXXX', ['finite element analysis', 'computer graphics'])\n", "('method used for task', 'the approach is based on a three-dimensional XXXXX and requires a low XXXXX', ['discrete element', 'computational cost'])\n", "('method used for task', 'the system is based on XXXXX and XXXXX', ['artificial neural networks', 'genetic algorithm'])\n", "('NONE', 'XXXXX for estimation of XXXXX', ['soft computing methods', 'wake wind speed'])\n", "('method used for task', 'propose the method of setting basic constraints to solve the shortcomings when XXXXX and haptic feedback are both integrated in assembly tasks . the basic constraints make the XXXXX easy and realistic through the usage of haptics and visual fidelity', ['physics engine', 'assembly training'])\n", "('NONE', 'except the display and XXXXX modules , the whole XXXXX is made up of closeable widgets', ['physics engine', 'training system'])\n", "('NONE', 'XXXXX of the XXXXX to mpi-only parallelization', ['performance comparison', 'hybrid parallelization'])\n", "('NONE', 'it is advisable among dentists to perform as diverse XXXXX as possible to reduce the risk of decreased XXXXX', ['work tasks', 'pinch grip strength'])\n", "('method used for task', 'delivering a first XXXXX for designing sustainable XXXXX', ['frame work', 'work systems'])\n", "('method used for task', 'a XXXXX function was derived to reveal the relationship between the muscle load and the XXXXX', ['subjective evaluation', 'satisfaction level'])\n", "('method used for task', 'a satisfaction level function was derived to reveal the relationship between the XXXXX and the XXXXX', ['subjective evaluation', 'muscle load'])\n", "('method used for task', 'a XXXXX function was derived to reveal the relationship between the XXXXX and the subjective evaluation', ['satisfaction level', 'muscle load'])\n", "('NONE', 'evaluation of on-road XXXXX was performed using XXXXX on 12 male volunteers', ['surface emg', 'bicycle design'])\n", "('method used for task', 'we formulate the relationship between perceived discomfort and joint XXXXX for the XXXXX', ['upper limb', 'moment ratio'])\n", "('method used for task', 'reconfiguration reduced XXXXX and enhanced XXXXX', ['error rates', 'user acceptance'])\n", "('method used for task', 'the seips model of XXXXX and XXXXX is a useful systems approach to healthcare quality and XXXXX', ['work system', 'patient safety'])\n", "('method used for task', 'the seips model of XXXXX and patient safety is a useful XXXXX to healthcare quality and patient safety', ['work system', 'systems approach'])\n", "('method used for task', 'the seips model of XXXXX and patient safety is a useful systems approach to XXXXX and patient safety', ['work system', 'healthcare quality'])\n", "('method used for task', 'the seips model of XXXXX and XXXXX is a useful systems approach to healthcare quality and XXXXX', ['work system', 'patient safety'])\n", "('method used for task', 'the seips model of work system and XXXXX is a useful XXXXX to healthcare quality and XXXXX', ['patient safety', 'systems approach'])\n", "('method used for task', 'the seips model of work system and XXXXX is a useful systems approach to XXXXX and XXXXX', ['patient safety', 'healthcare quality'])\n", "('method used for task', 'the seips model of work system and patient safety is a useful XXXXX to XXXXX and patient safety', ['systems approach', 'healthcare quality'])\n", "('method used for task', 'the seips model of work system and XXXXX is a useful XXXXX to healthcare quality and XXXXX', ['systems approach', 'patient safety'])\n", "('method used for task', 'the seips model of work system and XXXXX is a useful systems approach to XXXXX and XXXXX', ['healthcare quality', 'patient safety'])\n", "('method used for task', 'the seips model can be used for research and improvement activities for improving XXXXX and XXXXX', ['healthcare quality', 'patient safety'])\n", "('method used for task', 'the seips model can be used for research and XXXXX for improving XXXXX and patient safety', ['healthcare quality', 'improvement activities'])\n", "('method used for task', 'the seips model can be used for research and XXXXX for improving healthcare quality and XXXXX', ['patient safety', 'improvement activities'])\n", "('method used for task', 'balancing the XXXXX is a key principle to improve XXXXX and patient safety', ['work system', 'healthcare quality'])\n", "('method used for task', 'balancing the XXXXX is a key principle to improve healthcare quality and XXXXX', ['work system', 'patient safety'])\n", "('method used for task', 'balancing the work system is a key principle to improve XXXXX and XXXXX', ['healthcare quality', 'patient safety'])\n", "('NONE', 'an intergroup XXXXX identified distinct XXXXX of home heating that explained users reported behaviour', ['case study', 'mental models'])\n", "('NONE', 'an intergroup XXXXX identified distinct mental models of XXXXX that explained users reported behaviour', ['case study', 'home heating'])\n", "('NONE', 'an intergroup case study identified distinct XXXXX of XXXXX that explained users reported behaviour', ['mental models', 'home heating'])\n", "('NONE', 'electromyography , XXXXX , and ratings of XXXXX differentiated 3 hand-carried devices and a manual carry', ['heart rate', 'perceived exertion'])\n", "('method used for task', 'guidelines for developing a method for assessing all XXXXX and all XXXXX', ['work tasks', 'body parts'])\n", "('method used for task', 'wearing the XXXXX and s.c.b.a. , 14 % of all firefighters recruits failed to complete the XXXXX', ['protective clothing', 'endurance test'])\n", "('NONE', 'in a XXXXX ergonomic XXXXX are likely to be transformed , not just transferred', ['design process', 'guideline texts'])\n", "('NONE', \"XXXXX task 's ( lct ) sensitivity towards XXXXX was examined\", ['learning effects', 'lane change'])\n", "('method used for task', 'camouflage effectiveness was evaluated using hit rate , XXXXX , and XXXXX', ['detection time', 'eye movement data'])\n", "('NONE', 'results suggest that sa XXXXX showed some XXXXX', ['common ground', 'information elements'])\n", "('method used for task', 'trapezius XXXXX and workload were assessed over full day and XXXXX in nurses', ['muscle activity', 'night shifts'])\n", "('NONE', 'during day and XXXXX , XXXXX was very high compared to resting values', ['heart rate', 'night shifts'])\n", "('method used for task', 'perception of XXXXX and mental well-being at work were similar during day and XXXXX', ['neck pain', 'night shifts'])\n", "('NONE', 'XXXXX were perceived to be as burdensome as day shifts despite the smaller XXXXX', ['physical workload', 'night shifts'])\n", "('NONE', 'tested 14 universal XXXXX in the XXXXX , korea , and turkey', ['united states', 'healthcare symbols'])\n", "('NONE', 'poor health and XXXXX did not interact regarding reduced demand-specific XXXXX', ['working conditions', 'work ability'])\n", "('method used for task', 'XXXXX location varies by vehicle and XXXXX', ['hand placement', 'age group'])\n", "('NONE', 'effect of XXXXX predictability on XXXXX task ( lct ) performance was tested', ['lane change', 'lane task'])\n", "('NONE', 'effect of XXXXX predictability on XXXXX task ( lct ) performance was tested', ['lane change', 'lane task'])\n", "('NONE', 'finger flexor and extensor XXXXX was lower during the virtual XXXXX', ['muscle activity', 'keyboard use'])\n", "('NONE', 'a novel approach was developed to determine the XXXXX in XXXXX', ['air gap', 'protective clothing'])\n", "('method used for task', 'the XXXXX depended on the XXXXX , fabric properties , and garment size', ['air gap', 'body parts'])\n", "('NONE', 'neither system reduced core temperature , XXXXX , XXXXX or comfort', ['heart rate', 'perceived exertion'])\n", "('NONE', 'neither system reduced XXXXX , XXXXX , perceived exertion or comfort', ['heart rate', 'core temperature'])\n", "('NONE', 'neither system reduced XXXXX , heart rate , XXXXX or comfort', ['perceived exertion', 'core temperature'])\n", "('NONE', 'XXXXX in and around the firehouse may increase XXXXX', ['physical activity', 'body temperature'])\n", "('NONE', 'a list of XXXXX that influenced trust was derived from XXXXX', ['qualitative analysis', 'system features'])\n", "('NONE', 'parents and XXXXX members who reviewed video records of their bedside rounds participated in the analysis of XXXXX and facilitators in family-centered rounds', ['healthcare team', 'work system barriers'])\n", "('method used for task', 'the stimulated XXXXX was positively received by parents and XXXXX members', ['recall methodology', 'healthcare team'])\n", "('NONE', 'the stimulated XXXXX allowed the identification of a wide range of XXXXX and facilitators in family-centered rounds', ['recall methodology', 'work system barriers'])\n", "('NONE', 'XXXXX significantly lower on XXXXX due to decreased performance', ['muscle activity', 'slate computer'])\n", "('NONE', 'a toolkit approach is proposed to improve XXXXX of wmsds in XXXXX', ['risk management', 'health care'])\n", "('NONE', 'four different types of insoles were examined in terms of their effects on XXXXX in XXXXX', ['postural stability', 'older adults'])\n", "('NONE', 'the findings can aid in better understanding the insole XXXXX that could improve XXXXX in older adults', ['design features', 'postural stability'])\n", "('NONE', 'the findings can aid in better understanding the insole XXXXX that could improve postural stability in XXXXX', ['design features', 'older adults'])\n", "('NONE', 'the findings can aid in better understanding the insole design features that could improve XXXXX in XXXXX', ['postural stability', 'older adults'])\n", "('NONE', 'the effect of electric XXXXX on human heat balance during XXXXX was modelled', ['fan use', 'heat waves'])\n", "('NONE', 'we performed XXXXX of assembly training for automotive XXXXX', ['user studies', 'assembly lines'])\n", "('NONE', 'we performed XXXXX of XXXXX for automotive assembly lines', ['user studies', 'assembly training'])\n", "('method used for task', 'we performed user studies of XXXXX for automotive XXXXX', ['assembly lines', 'assembly training'])\n", "('method used for task', 'the study describes a longitudinal qualitative XXXXX examining efforts to improve behavioural XXXXX over a two- year period', ['case study', 'energy efficiency'])\n", "('method used for task', 'an association between prolonged XXXXX and XXXXX was found among women', ['arm elevation', 'shoulder pain'])\n", "('NONE', 'XXXXX increases carpal XXXXX in patients with carpal tunnel syndrome', ['computer use', 'tunnel pressure'])\n", "('method used for task', 'XXXXX times for XXXXX are slightly higher than for guide signs', ['visual processing', 'logo signs'])\n", "('method used for task', 'increased XXXXX times for XXXXX do not result in any vehicle control degradation', ['visual processing', 'logo signs'])\n", "('NONE', \"patients ' and informal caregivers ' XXXXX was shaped by XXXXX\", ['work performance', 'system factors'])\n", "('NONE', 'the XXXXX of the XXXXX has been calculated', ['human error probability', 'case system'])\n", "('NONE', 'some strategies to prevent latent XXXXX of the XXXXX have been proposed', ['human errors', 'case system'])\n", "('NONE', 'presentation of a personal XXXXX ( pcs ) XXXXX', ['cooling system', 'selection method'])\n", "('method used for task', 'the new XXXXX was superior to the current XXXXX in XXXXX and satisfaction score', ['error rate', 'touch method'])\n", "('method used for task', 'the new XXXXX was superior to the current XXXXX in XXXXX and satisfaction score', ['error rate', 'touch method'])\n", "('method used for task', 'XXXXX for the XXXXX have the least effect on the prediction', ['measurement errors', 'arc length'])\n", "('NONE', \"XXXXX of sports t-shirt increases wearer 's XXXXX\", ['thermal comfort', 'ventilation design'])\n", "('method used for task', 'an XXXXX on the human XXXXX inside the aircrafts was carried out', ['experimental investigation', 'thermal comfort'])\n", "('method used for task', 'appropriate ud XXXXX are varied on XXXXX', ['evaluation scales', 'product attributes'])\n", "('NONE', 'written XXXXX resulted in more accurate XXXXX than spoken channels', ['communication channels', 'message transmission'])\n", "('method used for task', 'high XXXXX forces and plantar pressures are observed during XXXXX at high gait cadences', ['load carriage', 'ground reaction'])\n", "('method used for task', 'the metabolic XXXXX was found to have a basic impact on the XXXXX', ['heat generation', 'temperature profile'])\n", "('method used for task', 'XXXXX was quantified and examined in relation to XXXXX', ['physical workload', 'strength capacities'])\n", "('NONE', 'XXXXX , XXXXX , trunk motion , and perceived exertion data were collected', ['surface emg', 'heart rate'])\n", "('method used for task', 'we present a practical approach based on XXXXX to estimate XXXXX using heart rate monitoring', ['neuro-fuzzy systems', 'energy expenditure'])\n", "('NONE', 'XXXXX texting requires less XXXXX than a physical keypad', ['touch screen', 'muscle activity'])\n", "('NONE', 'XXXXX demonstrates a more dynamic spine than XXXXX', ['keyboard use', 'mouse use'])\n", "('method used for task', 'speech-based agents enhanced driver XXXXX and XXXXX', ['situation awareness', 'driving performance'])\n", "('NONE', \"a driver 's XXXXX mediated anger effects on XXXXX\", ['situation awareness', 'driving performance'])\n", "('method used for task', 'we assess footwear XXXXX on ice by the maximum XXXXX subjects could stand and walk', ['slip resistance', 'slope angles'])\n", "('NONE', 'the maximum XXXXX were objective and ecologically valid measures of XXXXX', ['slip resistance', 'slope angles'])\n", "('NONE', \"this human-centred measure of XXXXX did not require controlling of individuals ' XXXXX\", ['slip resistance', 'gait characteristics'])\n", "('NONE', 'we conducted a XXXXX experiment under real XXXXX', ['lane change', 'road environment'])\n", "('NONE', 'XXXXX intent XXXXX is about 5 s', ['lane changing', 'time window'])\n", "('method used for task', 'vehicle motion states , driving conditions and XXXXX information were chosen to predict XXXXX behaviours', ['lane changing', 'head movements'])\n", "('NONE', 'the improved XXXXX detects 85 % of XXXXX 1.5 s in advance', ['neural network', 'lane changes'])\n", "('NONE', 'XXXXX that enriches how physicians view XXXXX', ['web-based training', 'diagnosis delivery'])\n", "('NONE', 'XXXXX provides XXXXX during blindfolded water pouring', ['motion capture', 'auditory feedback'])\n", "('method used for task', 'XXXXX and XXXXX are provided based on these results', ['system improvement', 'design recommendations'])\n", "('method used for task', 'we propose a novel hybrid XXXXX able to handle XXXXX sizes', ['small sample', 'regression method'])\n", "('NONE', 'outlines role of XXXXX in sustainability , and the role of XXXXX in these systems', ['cyber-physical systems', 'systems ergonomics'])\n", "('NONE', 'we examine XXXXX as a mediator in the relationships between job characteristics and XXXXX', ['job engagement', 'safety performance'])\n", "('method used for task', 'job resources had positive effects on XXXXX directly and indirectly through XXXXX', ['safety performance', 'job engagement'])\n", "('NONE', 'XXXXX can translate into better XXXXX', ['employee engagement', 'safety performance'])\n", "('NONE', 'XXXXX of XXXXX defined', ['objective measures', 'tool use'])\n", "('method used for task', 'XXXXX for site-specific schoolbag-related XXXXX were identified', ['risk factors', 'musculoskeletal discomfort'])\n", "('method used for task', 'we examine the influence of XXXXX and driving experience on subjective workload and XXXXX', ['driving performance', 'situation complexity'])\n", "('method used for task', 'we used XXXXX to analyse rail XXXXX', ['cognitive work analysis', 'level crossings'])\n", "('NONE', 'effects of individual and XXXXX on XXXXX were investigated', ['grip strength', 'job factors'])\n", "('NONE', 'opportunities exist to optimise XXXXX using XXXXX', ['virtual reality', 'vehicle development'])\n", "('NONE', 'the XXXXX reduced seat back XXXXX at the low back', ['thoracic support', 'contact pressure area'])\n", "('NONE', 'XXXXX played a leading causal role in clinical incidents , with XXXXX varying by type of action', ['human error', 'error type'])\n", "('NONE', 'it is suggested to evaluate the overall XXXXX of passengers using one single XXXXX ranging from extreme discomfort to extreme comfort', ['comfort experience', 'rating scale'])\n", "('method used for task', 'the XXXXX and XXXXX literatures are reviewed for insights regarding effective decision support design', ['decision making', 'problem solving'])\n", "('method used for task', 'the XXXXX and problem solving literatures are reviewed for insights regarding effective XXXXX', ['decision making', 'decision support design'])\n", "('method used for task', 'the decision making and XXXXX literatures are reviewed for insights regarding effective XXXXX', ['problem solving', 'decision support design'])\n", "('NONE', 'the role of graphical displays in supporting XXXXX , XXXXX and system safety is explored in detail', ['decision making', 'problem solving'])\n", "('method used for task', 'the role of graphical displays in supporting XXXXX , problem solving and XXXXX is explored in detail', ['decision making', 'system safety'])\n", "('method used for task', 'the role of graphical displays in supporting decision making , XXXXX and XXXXX is explored in detail', ['problem solving', 'system safety'])\n", "('NONE', 'participants prefer auditory warnings , as opposed to visual , vibrotactile or electric stimuli . specifically , XXXXX prefer alarm warnings , while XXXXX prefer verbal warnings', ['truck drivers', 'taxi drivers'])\n", "('NONE', 'XXXXX via the combination of technologies incorporating XXXXX , tri-axial accelerometers , and supported by survey data', ['data collection', 'gps data'])\n", "('NONE', 'XXXXX via the combination of technologies incorporating gps data , tri-axial accelerometers , and supported by XXXXX', ['data collection', 'survey data'])\n", "('method used for task', 'data collection via the combination of technologies incorporating XXXXX , tri-axial accelerometers , and supported by XXXXX', ['gps data', 'survey data'])\n", "('NONE', 'the maximum XXXXX and fatigue were influenced significantly in XXXXX and pressure', ['grip strength', 'low temperature'])\n", "('NONE', 'work sampling , XXXXX , XXXXX were integrated', ['computer simulation', 'biomechanical modeling'])\n", "('NONE', 'intrinsic XXXXX is an essential feature of XXXXX', ['human motion', 'movement variability'])\n", "('NONE', 'XXXXX theories can explain the existence of intrinsic XXXXX', ['motor control', 'movement variability'])\n", "('NONE', 'intrinsic XXXXX should be taken into account in XXXXX', ['movement variability', 'workstation design'])\n", "('method used for task', 'XXXXX and XXXXX in grocery store workers are clarified', ['physical workload', 'musculoskeletal symptoms'])\n", "('method used for task', 'XXXXX ( cwa ) is applied to pedestrian use of XXXXX', ['cognitive work analysis', 'rail level crossings'])\n", "('NONE', 'XXXXX was influenced by XXXXX but not by lay-off period', ['system reliability', 'operator trust'])\n", "('method used for task', 'XXXXX climate perceptions were linked to XXXXX and engagement', ['job satisfaction', 'employee safety'])\n", "('method used for task', 'XXXXX climate perceptions were linked to objective XXXXX', ['employee safety', 'turnover rate'])\n", "('NONE', 'XXXXX mediated between XXXXX , employee engagement , turnover rate', ['job satisfaction', 'safety climate'])\n", "('NONE', 'XXXXX mediated between safety climate , XXXXX , turnover rate', ['job satisfaction', 'employee engagement'])\n", "('NONE', 'XXXXX mediated between safety climate , employee engagement , XXXXX', ['job satisfaction', 'turnover rate'])\n", "('NONE', 'job satisfaction mediated between XXXXX , XXXXX , turnover rate', ['safety climate', 'employee engagement'])\n", "('NONE', 'job satisfaction mediated between XXXXX , employee engagement , XXXXX', ['safety climate', 'turnover rate'])\n", "('NONE', 'job satisfaction mediated between safety climate , XXXXX , XXXXX', ['employee engagement', 'turnover rate'])\n", "('method used for task', 'XXXXX should focus on the interactions between XXXXX and the role of higher level work system factors', ['future research', 'work systems'])\n", "('NONE', 'we examine whether XXXXX ( wbv ) increases frequency of XXXXX', ['whole-body vibration', 'action slips'])\n", "('method used for task', 'XXXXX times didn’t differ between 2d and 3d , but times were slower on the XXXXX than the microscope', ['task completion', 'video displays'])\n", "('method used for task', 'XXXXX reduce posture constraints and may reduce XXXXX and fatigue in microsurgery', ['musculoskeletal symptoms', 'video displays'])\n", "('NONE', 'the need of approaching and intervening simultaneously human and environmental problems has been identify . this are the results of a XXXXX of design concepts and methods associated with human and XXXXX', ['systematic review', 'environmental factors'])\n", "('NONE', 'the need of approaching and intervening simultaneously human and environmental problems has been identify . this are the results of a XXXXX of XXXXX and methods associated with human and environmental factors', ['systematic review', 'design concepts'])\n", "('method used for task', 'the need of approaching and intervening simultaneously human and environmental problems has been identify . this are the results of a systematic review of XXXXX and methods associated with human and XXXXX', ['environmental factors', 'design concepts'])\n", "('method used for task', 'it shows conceptual and methodological segregation between XXXXX and XXXXX in published documents', ['human factors', 'environmental factors'])\n", "('NONE', 'XXXXX was utilized to assess the improvement strategy of latent XXXXX', ['fuzzy topsis', 'human errors'])\n", "('NONE', 'XXXXX was used to inspect the robustness of the results of XXXXX', ['sensitivity analysis', 'fuzzy topsis'])\n", "('method used for task', 'this is the first study to assess XXXXX by hfacs and XXXXX', ['human errors', 'fuzzy topsis'])\n", "('method used for task', 'XXXXX are neither necessary nor sufficient to elicit XXXXX', ['time constraints', 'time pressure'])\n", "('method used for task', 'sorting quality influences XXXXX , environment and XXXXX', ['working conditions', 'system performance'])\n", "('method used for task', 'maritime engine room officers commonly agree that the rapid XXXXX and reduced staffing onboard have contributed to higher workload and altered XXXXX', ['technological development', 'work tasks'])\n", "('NONE', 'presented and provided evidence of how the XXXXX are categorised in each XXXXX', ['visualisation methods', 'selection approach'])\n", "('method used for task', 'neck disorders were associated with head and XXXXX , trapezius and forearm XXXXX , and wrist velocity', ['muscle activity', 'arm posture'])\n", "('NONE', 'workstations on XXXXX were classified into high and low workload workstations with the median score of XXXXX', ['assembly line', 'body parts'])\n", "('NONE', 'the hlcb could lower XXXXX , oxygen uptake , minute ventilation and peripheral rating of XXXXX when carrying a 15 kg load', ['heart rate', 'perceived exertion'])\n", "('NONE', 'XXXXX can address XXXXX ( e.g. , enhancing awareness )', ['human factors', 'policy interventions'])\n", "('NONE', 'the decision ladder was used to examine XXXXX at rail XXXXX', ['decision making', 'level crossings'])\n", "('NONE', 'XXXXX at rail XXXXX varies within and between road user groups', ['decision making', 'level crossings'])\n", "('NONE', 'XXXXX sonifications do not reveal when XXXXX is too high', ['pulse oximetry', 'oxygen saturation'])\n", "('NONE', 'pain reports of those with XXXXX on XXXXX decreased 58 % on decline', ['low back pain', 'level ground'])\n", "('NONE', 'during an instrument XXXXX proficiency test , XXXXX ( hr ) /heart rate variation ( hrv ) are sensitive measures of varying levels of pilot mental workload', ['heart rate', 'flight rules'])\n", "('NONE', 'this new approach provides more degrees of freedom and XXXXX in XXXXX', ['type-2 fuzzy logic systems', 'design flexibility'])\n", "('NONE', 'XXXXX using granular type-2 XXXXX have more potential to model and handle uncertainties', ['type-2 fuzzy logic systems', 'membership functions'])\n", "('NONE', 'XXXXX has been carried out without removing XXXXX in the data set', ['feature selection', 'missing values'])\n", "('method used for task', 'we show the application of gifss in real XXXXX and XXXXX', ['supplier selection', 'medical diagnosis problems'])\n", "('NONE', 'a matheuristic by combining the XXXXX with a XXXXX is suggested', ['tabu search algorithm', 'neighbourhood structure'])\n", "('NONE', 'we make use of XXXXX as a XXXXX for illumination classification', ['rough set', 'decision maker'])\n", "('method used for task', 'the evolving XXXXX is introduced for the modelling of XXXXX with dead-zone input', ['intelligent algorithm', 'nonlinear systems'])\n", "('NONE', 'applying the evolving computation concept to train a dynamic XXXXX capable of modeling XXXXX alloy actuators', ['black box', 'shape memory'])\n", "('NONE', 'we employ partially XXXXX for solving system of XXXXX', ['fuzzy neural network', 'fuzzy differential equations'])\n", "('NONE', 'we propose a XXXXX from the XXXXX for adjusting of weights', ['learning algorithm', 'cost function'])\n", "('NONE', 'XXXXX of open XXXXX and state of charge of lifepo4 batteries', ['mathematical model', 'circuit voltage'])\n", "('method used for task', 'use of XXXXX and fourier XXXXX for identifying similar past situations', ['hierarchical clustering', 'frequency analysis'])\n", "('method used for task', 'we use XXXXX to handle the XXXXX in the real-world problems', ['fuzzy set theory', 'imprecise information'])\n", "('method used for task', 'we use XXXXX to handle the imprecise information in the XXXXX', ['fuzzy set theory', 'real-world problems'])\n", "('NONE', 'we use fuzzy set theory to handle the XXXXX in the XXXXX', ['imprecise information', 'real-world problems'])\n", "('method used for task', 'a binary-real-coded XXXXX ( ga ) is proposed as the XXXXX of the ucp', ['genetic algorithm', 'solution technique'])\n", "('method used for task', 'we have considered in the XXXXX : continuous , integer and binary XXXXX simultaneously', ['optimization problem', 'decision variables'])\n", "('NONE', 'we introduce and examine a new methodology for training XXXXX ( rbf ) XXXXX', ['radial basis function', 'neural networks'])\n", "('NONE', 'we use XXXXX in order to improve the functionality of the optimum XXXXX ( osd ) learning algorithm', ['fuzzy clustering', 'steepest descent'])\n", "('NONE', 'we use XXXXX in order to improve the functionality of the optimum steepest descent ( osd ) XXXXX', ['fuzzy clustering', 'learning algorithm'])\n", "('NONE', 'we use fuzzy clustering in order to improve the functionality of the optimum XXXXX ( osd ) XXXXX', ['steepest descent', 'learning algorithm'])\n", "('method used for task', 'we can synthesize a set of XXXXX and find appropriate dithers to stabilize nonlinear multiple time-delay ( nmtd ) XXXXX', ['fuzzy controllers', 'interconnected systems'])\n", "('NONE', 'when the designed XXXXX can not stabilize the nmtd XXXXX , a batch of high-frequency signals ( commonly referred to as dithers ) is simultaneously introduced to stabilize it', ['fuzzy controllers', 'interconnected systems'])\n", "('NONE', 'gea-based pso can perform a XXXXX with faster XXXXX', ['global search', 'convergence speed'])\n", "('NONE', 'a hybrid optimization method for XXXXX of XXXXX in distribution networks is proposed', ['optimal allocation', 'wind turbines'])\n", "('method used for task', 'a XXXXX for XXXXX of wind turbines in distribution networks is proposed', ['optimal allocation', 'hybrid optimization method'])\n", "('NONE', 'a XXXXX for optimal allocation of XXXXX in distribution networks is proposed', ['wind turbines', 'hybrid optimization method'])\n", "('method used for task', 'combining the XXXXX ( ga ) and the market-based XXXXX ( opf )', ['genetic algorithm', 'optimal power flow'])\n", "('method used for task', 'diesel engine models are built using advanced XXXXX and verified based on XXXXX', ['machine learning techniques', 'experimental data'])\n", "('NONE', 'ntld classifier gives good XXXXX in case of XXXXX', ['classification accuracy', 'multi-class problem'])\n", "('method used for task', 'the solutions delivered by vn-mga are compared with the ones delivered by the XXXXX provided by an ilp solver , for medium-sized problem instances . the vn-mga was found to be able to find most of the exact XXXXX , leaving a small gap in the other cases', ['exact solutions', 'pareto-optimal solutions'])\n", "('NONE', 'the solutions delivered by vn-mga are compared with the ones delivered by the XXXXX provided by an XXXXX , for medium-sized problem instances . the vn-mga was found to be able to find most of the exact pareto-optimal solutions , leaving a small gap in the other cases', ['exact solutions', 'ilp solver'])\n", "('method used for task', 'the solutions delivered by vn-mga are compared with the ones delivered by the XXXXX provided by an ilp solver , for medium-sized XXXXX . the vn-mga was found to be able to find most of the exact pareto-optimal solutions , leaving a small gap in the other cases', ['exact solutions', 'problem instances'])\n", "('method used for task', 'the solutions delivered by vn-mga are compared with the ones delivered by the exact solutions provided by an XXXXX , for medium-sized problem instances . the vn-mga was found to be able to find most of the exact XXXXX , leaving a small gap in the other cases', ['pareto-optimal solutions', 'ilp solver'])\n", "('method used for task', 'the solutions delivered by vn-mga are compared with the ones delivered by the exact solutions provided by an ilp solver , for medium-sized XXXXX . the vn-mga was found to be able to find most of the exact XXXXX , leaving a small gap in the other cases', ['pareto-optimal solutions', 'problem instances'])\n", "('method used for task', 'the solutions delivered by vn-mga are compared with the ones delivered by the exact solutions provided by an XXXXX , for medium-sized XXXXX . the vn-mga was found to be able to find most of the exact pareto-optimal solutions , leaving a small gap in the other cases', ['ilp solver', 'problem instances'])\n", "('NONE', 'we present a supplier selection decision method based on XXXXX combined with a fuzzy rule-based XXXXX', ['fuzzy inference', 'classification method'])\n", "('method used for task', 'an adaptive hysteresis current XXXXX is proposed for dc–ac inverter in the XXXXX', ['control algorithm', 'pv system'])\n", "('NONE', 'we propose a network clustering algorithm , based on the application of XXXXX and capable of exploiting the XXXXX', ['genetic operators', 'traffic information'])\n", "('NONE', 'biological motivation , XXXXX , XXXXX , open research problems and challenging issues of these models', ['design principles', 'application areas'])\n", "('NONE', 'distributed XXXXX in a XXXXX including v2g', ['energy resources', 'smart grid environment'])\n", "('method used for task', 'an XXXXX named mofoa is proposed , which is the first study that extends the relatively new fireworks optimization heuristic for XXXXX', ['evolutionary algorithm', 'multiobjective optimization'])\n", "('NONE', 'XXXXX : modified shuffled frog leaping algorithm ( msfla ) with XXXXX ( ga ) cross-over', ['genetic algorithm', 'solution method'])\n", "('method used for task', 'an improved variant of the XXXXX is presented by simulating a XXXXX', ['abc algorithm', 'learning mechanism'])\n", "('NONE', 'indicating the XXXXX problems of the analytic XXXXX ( ahp ) , and proposing the paired interval scale addressing the limitations', ['rating scale', 'hierarchy process'])\n", "('NONE', 'a XXXXX combining XXXXX with wavelet radial basis function neural network is presented', ['hybrid algorithm', 'particle swarm optimization'])\n", "('NONE', 'a XXXXX combining particle swarm optimization with wavelet XXXXX is presented', ['hybrid algorithm', 'radial basis function neural network'])\n", "('NONE', 'a hybrid algorithm combining XXXXX with wavelet XXXXX is presented', ['particle swarm optimization', 'radial basis function neural network'])\n", "('NONE', 'development of evolutionary-tuned XXXXX for dynamic sequencing of jobs on a XXXXX', ['fuzzy controller', 'manufacturing facility'])\n", "('NONE', 'the proposed controllers produced considerably more XXXXX than the examined XXXXX', ['pareto optimal solutions', 'dispatching rules'])\n", "('method used for task', 'we propose XXXXX which is based on combining adaboost algorithm with svm for XXXXX phenomenon . we call this boosted svm', ['hybrid approach', 'imbalanced data'])\n", "('NONE', 'the proposed XXXXX minimizes weighted exponential XXXXX', ['hybrid approach', 'error function'])\n", "('NONE', 'the problem of constructing an organ XXXXX via XXXXX is assessed', ['machine learning techniques', 'allocation model'])\n", "('method used for task', 'the conclusion is based on the XXXXX and XXXXX ( i.e. , wilcoxon signed rank median test )', ['performance metrics', 'statistical analysis'])\n", "('method used for task', 'XXXXX for finite state machines is one of the main XXXXX in the synthesis of sequential circuits', ['state assignment', 'optimization problems'])\n", "('NONE', 'XXXXX for finite state machines is one of the main optimization problems in the synthesis of XXXXX', ['state assignment', 'sequential circuits'])\n", "('NONE', 'state assignment for finite state machines is one of the main XXXXX in the synthesis of XXXXX', ['optimization problems', 'sequential circuits'])\n", "('method used for task', 'an improved binary XXXXX is proposed and its effectiveness is demonstrated in solving the XXXXX', ['pso algorithm', 'state assignment problem'])\n", "('method used for task', 'one of the objective of XXXXX is to synthesize XXXXX targeting area optimization', ['sequential circuits', 'state assignment problem'])\n", "('NONE', 'an XXXXX using itae , XXXXX and settling times is proposed', ['objective function', 'damping ratio'])\n", "('NONE', 'an XXXXX for training the parameters of the XXXXX is further developed', ['optimal algorithm', 'forecasting model'])\n", "('method used for task', 'a XXXXX is examined to demonstrate the ability of the newly proposed XXXXX', ['case study', 'forecasting model'])\n", "('method used for task', 'the XXXXX approach is used to optimize local and global modelling capability of XXXXX', ['t-s fuzzy model', 'weighting parameters'])\n", "('NONE', 'the weighting parameters approach is used to optimize local and global XXXXX of XXXXX', ['t-s fuzzy model', 'modelling capability'])\n", "('method used for task', 'the XXXXX approach is used to optimize local and global XXXXX of t-s fuzzy model', ['weighting parameters', 'modelling capability'])\n", "('method used for task', 'proposed a XXXXX to classify the XXXXX images into two outcomes : benign or malignant', ['hybrid algorithm', 'breast cancer'])\n", "('NONE', 'the overall accuracy offered by the employed XXXXX confirms that the effectiveness and performance of the proposed XXXXX is high', ['hybrid technique', 'hybrid system'])\n", "('method used for task', 'the numerical simulation results confirm the superiority of the proposed ainet-sl in XXXXX and XXXXX', ['solution accuracy', 'convergence speed'])\n", "('NONE', 'we consider XXXXX in leader–follower formation with the XXXXX using the potential field method', ['path planning', 'obstacle avoidance'])\n", "('method used for task', 'a novel cost-sensitive ensemble , based on XXXXX , for XXXXX', ['decision trees', 'imbalanced classification'])\n", "('method used for task', 'XXXXX are utilized to solve multi-path XXXXX', ['meta-heuristic algorithms', 'selection problem'])\n", "('method used for task', 'hybrid oc–ga method has powerful capacity in searching for more optimal XXXXX and requiring less XXXXX', ['truss structures', 'computational effort'])\n", "('method used for task', 'XXXXX according to customer-specific XXXXX , which are highly relevant to the customers’ satisfaction level', ['vehicle routing', 'time windows'])\n", "('method used for task', 'XXXXX according to customer-specific time windows , which are highly relevant to the customers’ XXXXX', ['vehicle routing', 'satisfaction level'])\n", "('method used for task', 'vehicle routing according to customer-specific XXXXX , which are highly relevant to the customers’ XXXXX', ['time windows', 'satisfaction level'])\n", "('method used for task', 'the l1/2 penalized XXXXX is able to reduce the size of the predictor even further at moderate costs for the XXXXX', ['cox model', 'prediction accuracy'])\n", "('method used for task', 'the l1/2 penalized XXXXX is suitable for XXXXX with the high dimensional biological data', ['cox model', 'survival analysis'])\n", "('NONE', 'the user-interest ontology construction is proposed by using user log profile . we describe three steps of ontology construction approach : concept selection , the generation of optimized XXXXX , the process of translating XXXXX into user-interest ontology . it is worth to be mentioned that we can revise and update the user-interest ontology using the optimized XXXXX by fca . the more time the user uses XXXXX , the richer knowledge the user-interest ontology contains', ['concept lattice', 'search engine'])\n", "('NONE', 'the user-interest XXXXX is proposed by using user log profile . we describe three steps of XXXXX approach : concept selection , the generation of optimized XXXXX , the process of translating XXXXX into user-interest ontology . it is worth to be mentioned that we can revise and update the user-interest ontology using the optimized XXXXX by fca . the more time the user uses search engine , the richer knowledge the user-interest ontology contains', ['concept lattice', 'ontology construction'])\n", "('NONE', 'the user-interest XXXXX is proposed by using user log profile . we describe three steps of XXXXX approach : concept selection , the generation of optimized XXXXX , the process of translating XXXXX into user-interest ontology . it is worth to be mentioned that we can revise and update the user-interest ontology using the optimized XXXXX by fca . the more time the user uses search engine , the richer knowledge the user-interest ontology contains', ['concept lattice', 'ontology construction'])\n", "('NONE', 'the user-interest ontology construction is proposed by using user log profile . we describe three steps of ontology construction approach : concept selection , the generation of optimized XXXXX , the process of translating XXXXX into user-interest ontology . it is worth to be mentioned that we can revise and update the user-interest ontology using the optimized XXXXX by fca . the more time the user uses XXXXX , the richer knowledge the user-interest ontology contains', ['concept lattice', 'search engine'])\n", "('NONE', 'the user-interest XXXXX is proposed by using user log profile . we describe three steps of XXXXX approach : concept selection , the generation of optimized XXXXX , the process of translating XXXXX into user-interest ontology . it is worth to be mentioned that we can revise and update the user-interest ontology using the optimized XXXXX by fca . the more time the user uses search engine , the richer knowledge the user-interest ontology contains', ['concept lattice', 'ontology construction'])\n", "('NONE', 'the user-interest XXXXX is proposed by using user log profile . we describe three steps of XXXXX approach : concept selection , the generation of optimized XXXXX , the process of translating XXXXX into user-interest ontology . it is worth to be mentioned that we can revise and update the user-interest ontology using the optimized XXXXX by fca . the more time the user uses search engine , the richer knowledge the user-interest ontology contains', ['concept lattice', 'ontology construction'])\n", "('NONE', 'the user-interest ontology construction is proposed by using user log profile . we describe three steps of ontology construction approach : concept selection , the generation of optimized XXXXX , the process of translating XXXXX into user-interest ontology . it is worth to be mentioned that we can revise and update the user-interest ontology using the optimized XXXXX by fca . the more time the user uses XXXXX , the richer knowledge the user-interest ontology contains', ['concept lattice', 'search engine'])\n", "('NONE', 'the user-interest XXXXX is proposed by using user log profile . we describe three steps of XXXXX approach : concept selection , the generation of optimized XXXXX , the process of translating XXXXX into user-interest ontology . it is worth to be mentioned that we can revise and update the user-interest ontology using the optimized XXXXX by fca . the more time the user uses search engine , the richer knowledge the user-interest ontology contains', ['concept lattice', 'ontology construction'])\n", "('NONE', 'the user-interest XXXXX is proposed by using user log profile . we describe three steps of XXXXX approach : concept selection , the generation of optimized XXXXX , the process of translating XXXXX into user-interest ontology . it is worth to be mentioned that we can revise and update the user-interest ontology using the optimized XXXXX by fca . the more time the user uses search engine , the richer knowledge the user-interest ontology contains', ['concept lattice', 'ontology construction'])\n", "('NONE', 'the user-interest XXXXX is proposed by using user log profile . we describe three steps of XXXXX approach : concept selection , the generation of optimized concept lattice , the process of translating concept lattice into user-interest ontology . it is worth to be mentioned that we can revise and update the user-interest ontology using the optimized concept lattice by fca . the more time the user uses XXXXX , the richer knowledge the user-interest ontology contains', ['search engine', 'ontology construction'])\n", "('NONE', 'the user-interest XXXXX is proposed by using user log profile . we describe three steps of XXXXX approach : concept selection , the generation of optimized concept lattice , the process of translating concept lattice into user-interest ontology . it is worth to be mentioned that we can revise and update the user-interest ontology using the optimized concept lattice by fca . the more time the user uses XXXXX , the richer knowledge the user-interest ontology contains', ['search engine', 'ontology construction'])\n", "('NONE', 'based on user-interest ontology , we propose the seed urls selection approach . the user feature concept vectors , the XXXXX pages , the base set of hits , bipartite XXXXX , and the complete bipartite XXXXX are discussed for the seed urls selection based on user-interest ontology . the advantage of the approach is combing the semantic content and link information of web pages', ['semantic web', 'directed graph'])\n", "('NONE', 'based on user-interest ontology , we propose the seed urls selection approach . the user feature concept vectors , the XXXXX pages , the base set of hits , bipartite XXXXX , and the complete bipartite XXXXX are discussed for the seed urls selection based on user-interest ontology . the advantage of the approach is combing the semantic content and link information of web pages', ['semantic web', 'directed graph'])\n", "('method used for task', 'based on user-interest ontology , we propose the seed urls selection approach . the user feature concept vectors , the XXXXX pages , the base set of hits , bipartite directed graph , and the complete bipartite directed graph are discussed for the seed urls selection based on user-interest ontology . the advantage of the approach is combing the semantic content and link information of XXXXX', ['semantic web', 'web pages'])\n", "('NONE', 'based on user-interest ontology , we propose the seed urls XXXXX . the user feature concept vectors , the XXXXX pages , the base set of hits , bipartite directed graph , and the complete bipartite directed graph are discussed for the seed urls selection based on user-interest ontology . the advantage of the approach is combing the semantic content and link information of web pages', ['semantic web', 'selection approach'])\n", "('NONE', 'based on user-interest ontology , we propose the seed urls selection approach . the user feature concept vectors , the semantic XXXXX , the base set of hits , bipartite XXXXX , and the complete bipartite XXXXX are discussed for the seed urls selection based on user-interest ontology . the advantage of the approach is combing the semantic content and link information of XXXXX', ['directed graph', 'web pages'])\n", "('NONE', 'based on user-interest ontology , we propose the seed urls XXXXX . the user feature concept vectors , the semantic web pages , the base set of hits , bipartite XXXXX , and the complete bipartite XXXXX are discussed for the seed urls selection based on user-interest ontology . the advantage of the approach is combing the semantic content and link information of web pages', ['directed graph', 'selection approach'])\n", "('method used for task', 'based on user-interest ontology , we propose the seed urls selection approach . the user feature concept vectors , the semantic web pages , the base set of hits , bipartite XXXXX , and the complete bipartite XXXXX are discussed for the seed urls selection based on user-interest ontology . the advantage of the approach is combing the semantic content and link information of web pages', ['directed graph', 'feature vectors'])\n", "('NONE', 'based on user-interest ontology , we propose the seed urls selection approach . the user feature concept vectors , the semantic XXXXX , the base set of hits , bipartite XXXXX , and the complete bipartite XXXXX are discussed for the seed urls selection based on user-interest ontology . the advantage of the approach is combing the semantic content and link information of XXXXX', ['directed graph', 'web pages'])\n", "('NONE', 'based on user-interest ontology , we propose the seed urls XXXXX . the user feature concept vectors , the semantic web pages , the base set of hits , bipartite XXXXX , and the complete bipartite XXXXX are discussed for the seed urls selection based on user-interest ontology . the advantage of the approach is combing the semantic content and link information of web pages', ['directed graph', 'selection approach'])\n", "('method used for task', 'based on user-interest ontology , we propose the seed urls selection approach . the user feature concept vectors , the semantic web pages , the base set of hits , bipartite XXXXX , and the complete bipartite XXXXX are discussed for the seed urls selection based on user-interest ontology . the advantage of the approach is combing the semantic content and link information of web pages', ['directed graph', 'feature vectors'])\n", "('NONE', 'based on user-interest ontology , we propose the seed urls XXXXX . the user feature concept vectors , the semantic XXXXX , the base set of hits , bipartite directed graph , and the complete bipartite directed graph are discussed for the seed urls selection based on user-interest ontology . the advantage of the approach is combing the semantic content and link information of XXXXX', ['web pages', 'selection approach'])\n", "('method used for task', 'based on user-interest ontology , we propose the seed urls selection approach . the user feature concept vectors , the semantic XXXXX , the base set of hits , bipartite directed graph , and the complete bipartite directed graph are discussed for the seed urls selection based on user-interest ontology . the advantage of the approach is combing the semantic content and link information of XXXXX', ['web pages', 'feature vectors'])\n", "('method used for task', 'four fhr parameters are extracted from each fpcg signal . the XXXXX and XXXXX are designed based on standard guidelines', ['membership functions', 'fuzzy rules'])\n", "('method used for task', 'geometric optimisation , XXXXX and XXXXX are combined for achieving the above purposes', ['shape grammars', 'genetic algorithms'])\n", "('NONE', 'a semi-supervised XXXXX is proposed using modified XXXXX', ['self-organizing feature map', 'change detection method'])\n", "('NONE', 'we present a novel XXXXX on evolutionary XXXXX identifying the parameters of elm', ['model based', 'membrane algorithm'])\n", "('NONE', 'the classifiers are XXXXX , XXXXX , and nearest neighbor heuristics', ['logistic regression', 'decision tree'])\n", "('NONE', 'we present the results of two XXXXX real life XXXXX in the telco industry', ['large scale', 'case studies'])\n", "('NONE', 'system based on XXXXX , XXXXX and neural network', ['support vector machine', 'genetic algorithm'])\n", "('method used for task', 'system based on XXXXX , genetic algorithm and XXXXX', ['support vector machine', 'neural network'])\n", "('method used for task', 'system based on support vector machine , XXXXX and XXXXX', ['genetic algorithm', 'neural network'])\n", "('NONE', 'a comprehensive model is proposed for quick and accurate XXXXX and identification in XXXXX', ['fault detection', 'health monitoring'])\n", "('method used for task', 'a new method which combines the XXXXX and XXXXX is developed to classify data and accurately detect and isolate faults', ['k-means clustering', 'artificial neural networks'])\n", "('method used for task', 'we associate fqfd with relative XXXXX to identify adjusted XXXXX in an fmcdm model', ['preference relation', 'criteria weights'])\n", "('NONE', 'by the adjusted XXXXX , we can avoid multiplying triangular or XXXXX , aggregating multiplied fuzzy numbers , and ranking them', ['trapezoidal fuzzy numbers', 'criteria weights'])\n", "('method used for task', 'adjusted XXXXX substitutes for original XXXXX in fmcdm through relative XXXXX , and thus adjusted XXXXX are useful to construct an fmcdm model', ['preference relation', 'criteria weights'])\n", "('method used for task', 'adjusted XXXXX substitutes for original XXXXX in fmcdm through relative XXXXX , and thus adjusted XXXXX are useful to construct an fmcdm model', ['preference relation', 'criteria weights'])\n", "('method used for task', 'adjusted XXXXX substitutes for original XXXXX in fmcdm through relative XXXXX , and thus adjusted XXXXX are useful to construct an fmcdm model', ['preference relation', 'criteria weights'])\n", "('NONE', 'a novel nonlinear XXXXX for prediction of contact area and XXXXX', ['contact pressure', 'soft computing approach'])\n", "('method used for task', 'two anfis models have been designed to correlate the XXXXX to material removal rate ( mrr ) and XXXXX ( sr )', ['process parameters', 'surface roughness'])\n", "('method used for task', 'a continuous XXXXX ( caco ) technique has been used to select the best XXXXX for maximum mrr and specified sr', ['ant colony optimization', 'process parameters'])\n", "('method used for task', 'we incorporate XXXXX to create fuzzy XXXXX for helix pairs', ['fuzzy logic', 'contact maps'])\n", "('method used for task', 'retrieved XXXXX regions are used to predict a XXXXX using bound smoothing', ['distance map', 'distance map regions'])\n", "('method used for task', 'the embed algorithm is used on the predicted XXXXX to obtain XXXXX', ['distance map', '3d structure'])\n", "('method used for task', 'an lrgf XXXXX with pole assignment technique is proposed to model the XXXXX', ['neural network', 'dynamic system'])\n", "('method used for task', 'XXXXX are constructed to fit the XXXXX with every ɛ error', ['rbf neural networks', 'singular value'])\n", "('method used for task', 'XXXXX prediction is very important for XXXXX . in this paper , an intelligent approach is adopted in order to achieve XXXXX prediction and obtain good performance', ['software engineering', 'software reliability'])\n", "('method used for task', 'XXXXX is very important for XXXXX . in this paper , an intelligent approach is adopted in order to achieve XXXXX and obtain good performance', ['software engineering', 'software reliability prediction'])\n", "('method used for task', 'XXXXX is very important for XXXXX . in this paper , an intelligent approach is adopted in order to achieve XXXXX and obtain good performance', ['software engineering', 'software reliability prediction'])\n", "('method used for task', 'XXXXX prediction is very important for software engineering . in this paper , an intelligent approach is adopted in order to achieve XXXXX prediction and obtain good performance', ['software reliability', 'software reliability prediction'])\n", "('method used for task', 'XXXXX prediction is very important for software engineering . in this paper , an intelligent approach is adopted in order to achieve XXXXX prediction and obtain good performance', ['software reliability', 'software reliability prediction'])\n", "('NONE', 'pre-processing by XXXXX can greatly reduce XXXXX , and produces improved routings when the network nodes form aggregations', ['k-means clustering', 'computation time'])\n", "('NONE', 'with a goal to enhance the scalability and accuracy of XXXXX ( rss ) , we have introduced a new model which is a combination of both XXXXX and supervised learning', ['recommendation systems', 'unsupervised learning'])\n", "('NONE', 'with a goal to enhance the scalability and accuracy of XXXXX ( rss ) , we have introduced a new model which is a combination of both unsupervised learning and XXXXX', ['recommendation systems', 'supervised learning'])\n", "('method used for task', 'with a goal to enhance the scalability and accuracy of recommendation systems ( rss ) , we have introduced a new model which is a combination of both XXXXX and XXXXX', ['unsupervised learning', 'supervised learning'])\n", "('NONE', 'the feedforward and XXXXX of the XXXXX are self-tuned based on takagi-sugeno fuzzy model', ['feedback loops', 'control algorithm'])\n", "('NONE', 'the feedforward and XXXXX of the control algorithm are self-tuned based on takagi-sugeno XXXXX', ['feedback loops', 'fuzzy model'])\n", "('method used for task', 'the feedforward and feedback loops of the XXXXX are self-tuned based on takagi-sugeno XXXXX', ['control algorithm', 'fuzzy model'])\n", "('NONE', 'fast XXXXX with XXXXX generates the trajectory of each vehicle', ['genetic algorithm', 'bezier curve'])\n", "('method used for task', 'XXXXX are trained on-line with an XXXXX based algorithm', ['neural networks', 'extended kalman filter'])\n", "('NONE', 'XXXXX of XXXXX ( fcms ) with different functions are considered', ['fixed points', 'fuzzy cognitive maps'])\n", "('method used for task', 'we give a XXXXX for sigmoidal fcms to have unique stable XXXXX', ['sufficient condition', 'fixed points'])\n", "('method used for task', 'interval type-2 XXXXX ( it2fpid ) controllers are proposed for the XXXXX ( lfc ) problem', ['load frequency control', 'fuzzy pid'])\n", "('NONE', 'for the first time in literature , the big bang–big crunch algorithm is applied to tune the XXXXX and the footprint of uncertainty ( fou ) of XXXXX', ['membership functions', 'scaling factors'])\n", "('method used for task', 'the proposed optimum it2fpid load frequency controller is superior compared to the ordinary XXXXX and conventional XXXXX', ['pid controllers', 'fuzzy pid'])\n", "('method used for task', 'hw–sw partitioning of XXXXX is formulated as a XXXXX', ['embedded system', 'multi-objective problem'])\n", "('NONE', 'various XXXXX of the problem are taken as XXXXX', ['objective functions', 'cost terms'])\n", "('method used for task', 'a XXXXX repair technique for the adaptive XXXXX is proposed', ['crossover rate', 'de algorithm'])\n", "('method used for task', 'virtual screening methods can be improved using XXXXX and XXXXX', ['neural networks', 'support vector machines'])\n", "('method used for task', 'a new XXXXX called hybrid intelligence image fusion ( hiif ) is used for fusing multimodal XXXXX', ['hybrid algorithm', 'medical images'])\n", "('NONE', 'an application of fclarans for attribute clustering and XXXXX in XXXXX has been demonstrated', ['dimensionality reduction', 'gene expression data'])\n", "('method used for task', 'XXXXX based on XXXXX and differential gene expressions are employed in the process', ['domain knowledge', 'gene ontology'])\n", "('method used for task', 'the diagnosis of XXXXX ( cvds ) is faced using a linguistic fuzzy rule-based XXXXX', ['cardiovascular diseases', 'classification system'])\n", "('NONE', 'a multiple classifier system based on neural networks/support vector machines as base classifiers , balanced subspaces to address XXXXX , a XXXXX fuser and a fuzzy diversity measure is presented', ['class imbalance', 'neural network'])\n", "('method used for task', 'a multiple classifier system based on neural networks/support vector machines as XXXXX , balanced subspaces to address XXXXX , a neural network fuser and a fuzzy diversity measure is presented', ['class imbalance', 'base classifiers'])\n", "('method used for task', 'a multiple classifier system based on neural networks/support vector machines as XXXXX , balanced subspaces to address class imbalance , a XXXXX fuser and a fuzzy diversity measure is presented', ['neural network', 'base classifiers'])\n", "('method used for task', 'experimental results show excellent XXXXX on a challenging dataset and statistical superiority compared to other XXXXX', ['classification performance', 'ensemble classifiers'])\n", "('NONE', 'analyzing the performance of 75 XXXXX on a XXXXX', ['color constancy algorithms', 'benchmark dataset'])\n", "('NONE', 'this paper solves the real world problem of tactical XXXXX , in which optimality is sought to be achieved using the evolutionary computing method of XXXXX', ['differential evolution', 'missile guidance'])\n", "('NONE', 'we have applied an improved XXXXX sde , which is improved version of basic XXXXX ( de )', ['optimization technique', 'differential evolution'])\n", "('method used for task', 'it was adapted the fuzzy probability theory to the classical XXXXX for estimating the XXXXX of a component', ['reliability analysis', 'fuzzy reliability'])\n", "('NONE', 'the multi-center XXXXX can get rid of the problems that the XXXXX is sensitive to the initial prototypes , and it can handle non-traditional curved clusters', ['initialization method', 'fcm algorithm'])\n", "('method used for task', 'lpm-fft , XXXXX , and zernike moment are utilized to extract XXXXX', ['gabor filter', 'image features'])\n", "('NONE', 'the perceptual relativity has been applied to improve the performance of classification on the sparse , noisy or XXXXX , indicating the possibility of other perceptual laws in XXXXX being considered for classification', ['imbalanced data', 'cognitive psychology'])\n", "('NONE', 'reducing the XXXXX of optimization procedure of XXXXX using weighted least squares support vector machine', ['computational time', 'gravity dams'])\n", "('NONE', 'reducing the XXXXX of optimization procedure of gravity dams using XXXXX', ['computational time', 'weighted least squares support vector machine'])\n", "('NONE', 'reducing the computational time of optimization procedure of XXXXX using XXXXX', ['gravity dams', 'weighted least squares support vector machine'])\n", "('NONE', 'finding the XXXXX of concrete XXXXX with consideration of dam–water foundation rock interaction', ['optimal shape', 'gravity dams'])\n", "('method used for task', 'a hierarchical methodology based on the same principles of conventional takagi–sugeno identification ( XXXXX and XXXXX ) , that first determines the hybrid behavior of the system and then all the other non-linearities', ['principal components', 'fuzzy clustering'])\n", "('NONE', 'different XXXXX are compared and the impacts of parallel parameter on the proposed XXXXX are discussed', ['chaotic maps', 'hybrid algorithm'])\n", "('NONE', 'this paper proposes a self-organizing XXXXX as a XXXXX for ovarian cancer diagnoses', ['neural fuzzy system', 'decision support system'])\n", "('method used for task', 'XXXXX and XXXXX are performed during training', ['feature selection', 'attribute reduction'])\n", "('method used for task', 'we consider XXXXX and XXXXX to represent ambiguous , uncertain or imprecise information', ['fuzzy logic', 'fuzzy sets'])\n", "('method used for task', 'we consider XXXXX and fuzzy sets to represent ambiguous , uncertain or XXXXX', ['fuzzy logic', 'imprecise information'])\n", "('method used for task', 'we consider fuzzy logic and XXXXX to represent ambiguous , uncertain or XXXXX', ['fuzzy sets', 'imprecise information'])\n", "('method used for task', 'the integrated hybrid methodology is XXXXX and XXXXX for risk assessment', ['fuzzy ahp', 'fuzzy topsis'])\n", "('method used for task', 'the integrated hybrid methodology is XXXXX and fuzzy topsis for XXXXX', ['fuzzy ahp', 'risk assessment'])\n", "('method used for task', 'the integrated hybrid methodology is fuzzy ahp and XXXXX for XXXXX', ['fuzzy topsis', 'risk assessment'])\n", "('NONE', 'embeds XXXXX using a parameterized function that is automatically estimated using XXXXX', ['domain knowledge', 'stochastic optimization'])\n", "('method used for task', 'we propose a new type-2 XXXXX ( fs ) learned through structure and XXXXX', ['neural fuzzy system', 'parameter learning'])\n", "('method used for task', 'XXXXX and XXXXX are applied to markerless motion capture', ['evolutionary algorithms', 'particle filters'])\n", "('method used for task', 'XXXXX and particle filters are applied to markerless XXXXX', ['evolutionary algorithms', 'motion capture'])\n", "('method used for task', 'evolutionary algorithms and XXXXX are applied to markerless XXXXX', ['particle filters', 'motion capture'])\n", "('NONE', 'non-parametric tests show that XXXXX outperform XXXXX', ['evolutionary algorithms', 'particle filters'])\n", "('NONE', 'XXXXX and hierarchical strategies deal better with XXXXX', ['evolutionary algorithms', 'high dimensionality'])\n", "('method used for task', 'we present our versions of two well-known XXXXX ( nsgaii and spea2 ) applied to this XXXXX', ['multiobjective evolutionary algorithms', 'mobile network problem'])\n", "('method used for task', 'feature-level and XXXXX is explored using svm and rf classifiers , wrapper-based amde XXXXX is also investigated', ['decision-level fusion', 'feature selection'])\n", "('method used for task', 'feature-level and XXXXX is explored using svm and XXXXX , wrapper-based amde feature selection is also investigated', ['decision-level fusion', 'rf classifiers'])\n", "('NONE', 'feature-level and decision-level fusion is explored using svm and XXXXX , wrapper-based amde XXXXX is also investigated', ['feature selection', 'rf classifiers'])\n", "('method used for task', 'we contribute to XXXXX decomposition approaches by introducing a simple feature oriented technique for ensemble design , which is based on partitioning the XXXXX by XXXXX', ['feature set', 'k-means clustering'])\n", "('NONE', 'results indicate that ensemble of XXXXX , induced using the proposed technique , significantly improve decision-level and outperform XXXXX', ['feature-level fusion', 'rf classifiers'])\n", "('NONE', 'a set of three evolving XXXXX are derived by an online XXXXX', ['tsk fuzzy models', 'identification algorithm'])\n", "('method used for task', 'an improved fkp-mco classifier based on XXXXX , kernel technique , and XXXXX is proposed and is used for predicting protein–protein interaction hot spots', ['fuzzification method', 'penalty factors'])\n", "('NONE', 'the XXXXX uses XXXXX and moving average technical index to predict stock price trends', ['hybrid model', 'linear model'])\n", "('NONE', 'a XXXXX is proposed for selection of abstract data types implementations during the execution of a XXXXX', ['support vector machine', 'software system'])\n", "('NONE', 'the considered XXXXX confirm a good performance of the proposed model and show that using XXXXX for the data selection problem is worth being further investigated', ['computational experiments', 'machine learning techniques'])\n", "('method used for task', 'this study proposes a XXXXX based on the fuzzy decision-making trail and evaluation laboratory ( dematel ) and fuzzy multi-criteria decision-making ( fmcdm ) for XXXXX in sc', ['prediction framework', 'km adoption'])\n", "('NONE', 'it also enables organizations to decide whether to initiate XXXXX , restrain adoption or undertake remedial improvements to increase the possibility of successful XXXXX in sc', ['knowledge management', 'km adoption'])\n", "('NONE', 'XXXXX acts as XXXXX for ga', ['support vector machine', 'objective function'])\n", "('NONE', 'mlp is considered for each XXXXX for accurate estimation of dry XXXXX', ['soil types', 'unit weights'])\n", "('method used for task', 'a new pso-based XXXXX is presented to solve dynamic economic dispatch problems with XXXXX', ['hybrid algorithm', 'valve-point effects'])\n", "('method used for task', 'we propose a novel XXXXX for applicability of hyper-heuristic techniques on XXXXX', ['hybrid strategy', 'dynamic environments'])\n", "('method used for task', 'performance of our method is validated with the dynamic XXXXX and the moving XXXXX', ['generalized assignment problem', 'peaks benchmark'])\n", "('NONE', 'point out the drawbacks of the XXXXX and the parameter settings of XXXXX ( de )', ['crossover operator', 'differential evolution'])\n", "('method used for task', 'we propose a fast XXXXX to uncover XXXXX in networks', ['memetic algorithm', 'community structure'])\n", "('method used for task', 'to the best of our knowledge this is the first attempt to apply the XXXXX to a real-industrial XXXXX', ['cuckoo search algorithm', 'scheduling problem'])\n", "('method used for task', 'this paper proposes a new XXXXX for solving XXXXX', ['heuristic algorithm', 'optimization problems'])\n", "('method used for task', 'this XXXXX is inspired of trading the shares on XXXXX', ['optimization algorithm', 'stock market'])\n", "('NONE', 'pitfalls in the comparison of XXXXX within the field of applied XXXXX are quite common', ['computational experiments', 'evolutionary computing'])\n", "('method used for task', 'common pitfalls found in replication and comparison of XXXXX for economic XXXXX problem are presented', ['computational experiments', 'load dispatch'])\n", "('method used for task', 'guidelines on setting and conducting XXXXX for XXXXX are provided', ['computational experiments', 'evolutionary algorithms'])\n", "('NONE', 'XXXXX with a set of large-scale instances show that the mdsfl can be an efficient alternative for solving tightly constrained 01 XXXXX', ['computational experiments', 'knapsack problems'])\n", "('method used for task', 'the partition based clustering algorithms k-means and XXXXX algorithms are taken for analysis via its XXXXX', ['fuzzy c-means', 'computational time'])\n", "('method used for task', 'the distribution of data points by XXXXX is even to all the XXXXX , but , it is not even by the fcm algorithm', ['k-means algorithm', 'data centers'])\n", "('NONE', 'the distribution of XXXXX by XXXXX is even to all the data centers , but , it is not even by the fcm algorithm', ['k-means algorithm', 'data points'])\n", "('method used for task', 'the distribution of data points by XXXXX is even to all the data centers , but , it is not even by the XXXXX', ['k-means algorithm', 'fcm algorithm'])\n", "('method used for task', 'the distribution of XXXXX by k-means algorithm is even to all the XXXXX , but , it is not even by the fcm algorithm', ['data centers', 'data points'])\n", "('method used for task', 'the distribution of data points by k-means algorithm is even to all the XXXXX , but , it is not even by the XXXXX', ['data centers', 'fcm algorithm'])\n", "('method used for task', 'the distribution of XXXXX by k-means algorithm is even to all the data centers , but , it is not even by the XXXXX', ['data points', 'fcm algorithm'])\n", "('NONE', 'from the XXXXX , the XXXXX of k-means algorithm is less than the fcm algorithm', ['experimental analysis', 'computational time'])\n", "('NONE', 'from the XXXXX , the computational time of XXXXX is less than the fcm algorithm', ['experimental analysis', 'k-means algorithm'])\n", "('NONE', 'from the experimental analysis , the XXXXX of XXXXX is less than the fcm algorithm', ['computational time', 'k-means algorithm'])\n", "('NONE', 'an elitist self-adaptive step-size search ( esass ) algorithm is proposed for XXXXX of XXXXX subject to stress and displacement constraints', ['optimum design', 'truss structures'])\n", "('NONE', 'the XXXXX of the technique is accelerated through avoiding unnecessary analyses throughout the XXXXX using the so-called upper bound strategy ( ubs )', ['computational efficiency', 'optimization process'])\n", "('NONE', 'the esass algorithm is capable of locating reasonable solutions to optimum sizing problems of XXXXX with considerably less XXXXX', ['truss structures', 'computational effort'])\n", "('method used for task', 'XXXXX probability is adapted based on effectiveness of XXXXX operator', ['local search', 'local search operator'])\n", "('method used for task', 'maximum profit has been compared by XXXXX and XXXXX based XXXXX', ['genetic algorithm', 'fuzzy simulation'])\n", "('method used for task', 'maximum profit has been compared by XXXXX and XXXXX based XXXXX', ['fuzzy simulation', 'genetic algorithm'])\n", "('method used for task', 'the concept of the XXXXX is used to collect the real-time XXXXX and derives the appropriate paths for users', ['cellular automata', 'road conditions'])\n", "('NONE', 'the proposed work makes use of the XXXXX and the concept of the XXXXX to reduce the computational complexity', ['hierarchical structure', 'cellular automata'])\n", "('method used for task', 'the proposed work makes use of the XXXXX and the concept of the cellular automata to reduce the XXXXX', ['hierarchical structure', 'computational complexity'])\n", "('method used for task', 'the proposed work makes use of the hierarchical structure and the concept of the XXXXX to reduce the XXXXX', ['cellular automata', 'computational complexity'])\n", "('NONE', 'fuzzy XXXXX ( fsd ) intertwines XXXXX and XXXXX', ['statistical downscaling', 'fuzzy system'])\n", "('NONE', 'fuzzy XXXXX ( fsd ) intertwines XXXXX and XXXXX', ['fuzzy system', 'statistical downscaling'])\n", "('NONE', 'the XXXXX can be executed rapidly or with XXXXX by even systems based on fixed-point processors', ['control algorithm', 'low power consumption'])\n", "('method used for task', 'develops five XXXXX for parallel-machine and inbound-trucks sequencing in multi door XXXXX', ['hybrid metaheuristics', 'cross docking'])\n", "('method used for task', 'the hybrid simulated-annealing tabu-search algorithm is the best if XXXXX is used the XXXXX instead', ['cpu time', 'stopping criterion'])\n", "('NONE', 'aim at preventing overfitting of the state of the XXXXX ( XXXXX ) and improves upon other ( indictive method )', ['least squares', 'art methods'])\n", "('method used for task', 'adaptive XXXXX have been constructed in the proposed technique for both bell-shaped and triangle-shaped XXXXX', ['fuzzy membership functions', 'fuzzy sets'])\n", "('NONE', 'idea of multiple XXXXX has been incorporated in an effective manner for the removal of XXXXX from degraded images', ['fuzzy membership functions', 'impulse noise'])\n", "('method used for task', 'detailed XXXXX is presented , showing that different XXXXX perform better for different types of images at different noise ratio', ['sensitivity analysis', 'fuzzy membership functions'])\n", "('method used for task', 'detailed XXXXX is presented , showing that different fuzzy membership functions perform better for different types of images at different XXXXX', ['sensitivity analysis', 'noise ratio'])\n", "('NONE', 'detailed sensitivity analysis is presented , showing that different XXXXX perform better for different types of images at different XXXXX', ['fuzzy membership functions', 'noise ratio'])\n", "('method used for task', 'performance results based on the igd metric show that the XXXXX is overall superior to versions with single XXXXX', ['hybrid algorithm', 'search operators'])\n", "('method used for task', 'the proposed method is experimented on 13 unconstrained XXXXX and 24 constrained XXXXX , 20 XXXXX and 22 real life problems from the literature', ['benchmark problems', 'mechanical design problems'])\n", "('method used for task', 'the proposed method is experimented on 13 unconstrained XXXXX and 24 constrained XXXXX , 20 XXXXX and 22 real life problems from the literature', ['benchmark problems', 'mechanical design problems'])\n", "('method used for task', 'describes an XXXXX and a multi-round XXXXX for solving the problem', ['heuristic algorithm', 'integer programming formulation'])\n", "('NONE', 'the paper contributes to support the hypothesis that hybrid and XXXXX can be used in XXXXX without necessarily incurring significantly higher distance-based costs', ['electric vehicles', 'routing problems'])\n", "('NONE', 'compares XXXXX and fuzzy ahp methods concerning the problem of XXXXX based on a set of seven criteria', ['fuzzy topsis', 'supplier selection'])\n", "('NONE', 'further work can explore alternative approaches to avoid nulling weights of the criteria and XXXXX in XXXXX', ['rank reversal', 'fuzzy ahp'])\n", "('NONE', 'an adaptive hybrid control system using rrsefnn for pmsm servo drive is proposed.the computed torque control is designed to stabilize the pmsm servo drive.the rrsefnn is used to estimate the nonlinear lumped uncertainty of the pmsm drive.the online adaptive laws are derived using the XXXXX analysis.all XXXXX are implemented using tms320c31 dsp-based control computer', ['lyapunov stability', 'control algorithms'])\n", "('method used for task', 'an adaptive hybrid control system using rrsefnn for pmsm servo drive is proposed.the computed XXXXX is designed to stabilize the pmsm servo drive.the rrsefnn is used to estimate the nonlinear lumped uncertainty of the pmsm drive.the online adaptive laws are derived using the XXXXX analysis.all control algorithms are implemented using tms320c31 dsp-based control computer', ['lyapunov stability', 'torque control'])\n", "('NONE', 'an adaptive hybrid control system using rrsefnn for pmsm servo drive is proposed.the computed torque control is designed to stabilize the pmsm servo drive.the rrsefnn is used to estimate the nonlinear lumped uncertainty of the pmsm drive.the online adaptive laws are derived using the XXXXX analysis.all control algorithms are implemented using tms320c31 dsp-based XXXXX', ['lyapunov stability', 'control computer'])\n", "('method used for task', 'an adaptive hybrid control system using rrsefnn for pmsm servo drive is proposed.the computed XXXXX is designed to stabilize the pmsm servo drive.the rrsefnn is used to estimate the nonlinear lumped uncertainty of the pmsm drive.the online adaptive laws are derived using the lyapunov stability analysis.all XXXXX are implemented using tms320c31 dsp-based control computer', ['control algorithms', 'torque control'])\n", "('NONE', 'an adaptive hybrid control system using rrsefnn for pmsm servo drive is proposed.the computed torque control is designed to stabilize the pmsm servo drive.the rrsefnn is used to estimate the nonlinear lumped uncertainty of the pmsm drive.the online adaptive laws are derived using the lyapunov stability analysis.all XXXXX are implemented using tms320c31 dsp-based XXXXX', ['control algorithms', 'control computer'])\n", "('method used for task', 'an adaptive hybrid control system using rrsefnn for pmsm servo drive is proposed.the computed XXXXX is designed to stabilize the pmsm servo drive.the rrsefnn is used to estimate the nonlinear lumped uncertainty of the pmsm drive.the online adaptive laws are derived using the lyapunov stability analysis.all control algorithms are implemented using tms320c31 dsp-based XXXXX', ['torque control', 'control computer'])\n", "('NONE', 'proposed system used ensemble XXXXX with svm as XXXXX for the classification of XXXXX as normal or tumor', ['brain images', 'base classifier'])\n", "('NONE', 'proposed system used ensemble XXXXX with svm as XXXXX for the classification of XXXXX as normal or tumor', ['brain images', 'base classifier'])\n", "('method used for task', 'XXXXX based weighted majority voting scheme has been used for XXXXX', ['genetic algorithm', 'ensemble classifiers'])\n", "('NONE', 'the problem of finding the expected XXXXX in XXXXX has numerous applications', ['shortest path', 'stochastic networks'])\n", "('NONE', 'the proposed scheme is subjected to some XXXXX as well as XXXXX suites', ['security analysis', 'statistical tests'])\n", "('method used for task', 'we propose a XXXXX for XXXXX', ['human resources', 'performance evaluation method'])\n", "('NONE', 'the result shows that the proposed algorithm can generate better XXXXX within shorter XXXXX and stable convergence characteristics compared to fapso , pso and ga', ['computational time', 'quality solution'])\n", "('method used for task', 'the result shows that the proposed algorithm can generate better quality solution within shorter XXXXX and stable XXXXX compared to fapso , pso and ga', ['computational time', 'convergence characteristics'])\n", "('method used for task', 'the result shows that the proposed algorithm can generate better XXXXX within shorter computational time and stable XXXXX compared to fapso , pso and ga', ['quality solution', 'convergence characteristics'])\n", "('method used for task', 'existing XXXXX detection techniques are classifying the given signal as normal voice and XXXXX . but no technique is to detect the type of pathology ( analyzed the techniques published during last five years )', ['pathological voice', 'voice pathology'])\n", "('method used for task', 'a two level system has been proposed for detecting the type of XXXXX . in the first level , the system identifies whether the given voice is normal or not . if it is recognized as XXXXX , in the second level which aims at identifying the particular type of pathology', ['pathological voice', 'voice pathology'])\n", "('NONE', 'a two-level model is proposed for real-time XXXXX in a XXXXX', ['path planning', 'dynamic environment'])\n", "('NONE', 'in case of XXXXX the lack of XXXXX of small magnitude is only occasionally observed', ['benchmark problems', 'difference vectors'])\n", "('method used for task', 'the algorithm is tested using XXXXX and a XXXXX', ['benchmark functions', 'real-world application'])\n", "('method used for task', 'XXXXX was used to generate the XXXXX', ['subtractive clustering', 'membership functions'])\n", "('NONE', 'XXXXX allowed training the anfis with XXXXX with noise', ['subtractive clustering', 'experimental data'])\n", "('NONE', 'results illustrate that the fuzzy calculator is an efficient tool to propagate XXXXX regarding non-interactive as well as interactive fuzzy XXXXX', ['epistemic uncertainty', 'input variables'])\n", "('method used for task', 'in this type of strategy to combine the potential for aggregation of information/knowledge of the XXXXX for processing XXXXX in the bayesian network simplicity we will call this combination of fuzzy/bayesian network', ['fuzzy sets theory', 'input uncertainties'])\n", "('NONE', 'to illustrate the efficiency of the proposed methodology , the problem of XXXXX in the stator winding of XXXXX is presented', ['fault detection', 'induction machine'])\n", "('method used for task', 'the new strategy is based on the potential for aggregation of information/knowledge the XXXXX for processing input uncertainties in the XXXXX', ['fuzzy sets theory', 'bayesian network'])\n", "('method used for task', 'the new strategy is based on the potential for aggregation of information/knowledge the XXXXX for processing XXXXX in the bayesian network', ['fuzzy sets theory', 'input uncertainties'])\n", "('NONE', 'the new strategy is based on the potential for aggregation of information/knowledge the fuzzy sets theory for processing XXXXX in the XXXXX', ['bayesian network', 'input uncertainties'])\n", "('NONE', 'we propose a new XXXXX for segmentation of XXXXX', ['hybrid model', 'sar images'])\n", "('NONE', 'XXXXX captures the textural information of XXXXX more precisely', ['neutrosophic set', 'sar image'])\n", "('NONE', 'the proposed method can segment the XXXXX containing XXXXX', ['sar images', 'speckle noise'])\n", "('method used for task', 'the XXXXX is formulated as a XXXXX', ['multi-objective optimization problem', 'optimal power flow problem'])\n", "('method used for task', 'modifying the XXXXX by applying opposition based learning ( obl ) to enhance its XXXXX', ['tlbo algorithm', 'search ability'])\n", "('NONE', 'we propose XXXXX based XXXXX to train the proposed anns', ['genetic algorithm', 'learning algorithm'])\n", "('method used for task', 'fuzzy analytical XXXXX is efficient to extract customers’ preferences for XXXXX', ['hierarchy process', 'core attributes'])\n", "('NONE', 'we propose a set of XXXXX to formally reduce the problem of XXXXX pctlc to the problem of XXXXX pctl', ['reduction rules', 'model checking'])\n", "('method used for task', 'the application of XXXXX and biogeography based optimization to generate near-optimal XXXXX is being proposed', ['genetic algorithm', 'golomb ruler sequences'])\n", "('NONE', 'both the approaches produce near-optimal XXXXX very efficiently in reasonable XXXXX', ['execution time', 'golomb ruler sequences'])\n", "('NONE', 'in particular , there have been recent applications of XXXXX in the fields of XXXXX , classification and clustering , where it has helped improving results over type-1 fuzzy logic', ['type-2 fuzzy logic', 'pattern recognition'])\n", "('method used for task', 'since sample data may include uncertainties coming from measurement systems and environmental conditions , XXXXX and/or XXXXX can be used to capture these uncertainties', ['fuzzy numbers', 'linguistic variables'])\n", "('method used for task', 'since XXXXX may include uncertainties coming from measurement systems and environmental conditions , XXXXX and/or linguistic variables can be used to capture these uncertainties', ['fuzzy numbers', 'sample data'])\n", "('method used for task', 'since XXXXX may include uncertainties coming from measurement systems and environmental conditions , fuzzy numbers and/or XXXXX can be used to capture these uncertainties', ['linguistic variables', 'sample data'])\n", "('NONE', 'in this paper , one of the most popular XXXXX , exponentially weighted moving average XXXXX ( ewma ) for univariate data are developed under fuzzy environment', ['control charts', 'control chart'])\n", "('method used for task', 'in this paper , one of the most popular XXXXX , exponentially weighted moving average control chart ( ewma ) for univariate data are developed under XXXXX', ['control charts', 'fuzzy environment'])\n", "('method used for task', 'in this paper , one of the most popular control charts , exponentially weighted moving average XXXXX ( ewma ) for univariate data are developed under XXXXX', ['control chart', 'fuzzy environment'])\n", "('NONE', 'automated retinal vessel segmentation technique using XXXXX for screening of XXXXX', ['neural network', 'diabetic retinopathy'])\n", "('NONE', 'proposed method has XXXXX of the vasculature in XXXXX', ['automatic recognition', 'retinal images'])\n", "('NONE', 'we compare som and ghsom models based on the XXXXX and goodness of XXXXX to find the efficacy of the performance on a set of 15 benchmarked problems', ['network architecture', 'cell formation'])\n", "('NONE', 'i propose a growing XXXXX developed from perspective of a XXXXX', ['neural network', 'generative model'])\n", "('NONE', 'surveying various type-2 fuzzy disciplines including XXXXX , XXXXX , etc', ['fuzzy systems', 'fuzzy clustering'])\n", "('method used for task', 'this paper presents an effective biased random-key XXXXX for the tactical XXXXX', ['genetic algorithm', 'berth allocation problem'])\n", "('method used for task', 'correlation between grasping XXXXX and embedded XXXXX', ['objects weight', 'sensor stress'])\n", "('method used for task', 'XXXXX is performed using two XXXXX', ['parametric identification', 'soft computing techniques'])\n", "('method used for task', 'good agreement is found between XXXXX and XXXXX', ['experimental data', 'numerical simulations'])\n", "('method used for task', 'new email spam detection model based on XXXXX and XXXXX ( nsa–pso ) is implemented', ['negative selection algorithm', 'particle swarm optimization'])\n", "('NONE', 'random detector generation is replaced with XXXXX ; XXXXX and threshold value were studied to select distinctive features for spam detection', ['particle swarm optimization', 'distance measure'])\n", "('NONE', 'this paper proposes a multi-objective XXXXX of type-2 XXXXX', ['genetic optimization', 'fuzzy controllers'])\n", "('method used for task', 'the adaptive relevance vector machine is compared to the XXXXX and the XXXXX in the estimation', ['artificial neural networks', 'support vector machine'])\n", "('method used for task', 'the adaptive XXXXX machine is compared to the XXXXX and the support vector machine in the estimation', ['artificial neural networks', 'relevance vector'])\n", "('method used for task', 'the adaptive XXXXX machine is compared to the artificial neural networks and the XXXXX in the estimation', ['support vector machine', 'relevance vector'])\n", "('NONE', 'a transformation mechanism including three XXXXX between intuitionistic XXXXX and intuitionistic multiplicative preference relations are established', ['fuzzy preference relations', 'transformation functions'])\n", "('NONE', 'a new multi-objective method based on thd , htll , mllf , and total apfs currents to eliminate XXXXX in the electrical XXXXX is taken into account', ['harmonic distortion', 'distribution network'])\n", "('method used for task', 'two XXXXX from literature are studied and the results illustrate significant improvement in structural weight compared to those of the conventional XXXXX', ['numerical examples', 'design methods'])\n", "('NONE', 'a novel XXXXX of hybrid neuro-fuzzy for XXXXX of four area power system is presented', ['load frequency control', 'control approach'])\n", "('method used for task', 'XXXXX is carried out by using fuzzy , ann , anfis and conventional pi and XXXXX approaches', ['performance evaluation', 'pid control'])\n", "('NONE', 'single-chip embedded mimo autonomous XXXXX with on-line learning capability : small footprint , XXXXX , transparent device', ['intelligent agent', 'low power'])\n", "('NONE', 'we know the XXXXX of the XXXXX affecting a component', ['stochastic model', 'degradation process'])\n", "('NONE', 'we develop a method based on XXXXX of evidence and XXXXX to propagate the uncertainty', ['dempster–shafer theory', 'fuzzy random variables'])\n", "('method used for task', 'only XXXXX is applied , but it can reach a high security and less XXXXX', ['diffusion function', 'time cost'])\n", "('NONE', 'we examined the XXXXX on sunspots , XXXXX and indian stock data', ['model accuracy', 'electricity price'])\n", "('method used for task', 'binary XXXXX ( bgsa ) is proposed to solve XXXXX', ['gravitational search algorithm', 'unit commitment'])\n", "('NONE', 'we developed an XXXXX ( ann ) model to mimic route choice behaviour in crowds which achieved a XXXXX of 86 %', ['artificial neural network', 'prediction accuracy'])\n", "('method used for task', 'a XXXXX is proposed to make a XXXXX , called as physarum algorithm , more efficient', ['heuristic method', 'bio-inspired algorithm'])\n", "('method used for task', 'XXXXX is employed for the determination of XXXXX', ['pso algorithm', 'model parameters'])\n", "('NONE', \"an orthogonal forward selection algorithm is proposed for constructing radial basis function classifiers based on maximises the leave-one-out XXXXX between the classifier 's predicted XXXXX and the true XXXXX\", ['mutual information', 'class labels'])\n", "('NONE', \"an orthogonal forward selection algorithm is proposed for constructing radial basis function classifiers based on maximises the leave-one-out XXXXX between the classifier 's predicted XXXXX and the true XXXXX\", ['mutual information', 'class labels'])\n", "('method used for task', 'a novel XXXXX for XXXXX ( ba ) has been proposed', ['bees algorithm', 'initialization algorithm'])\n", "('NONE', 'a dead-zone XXXXX can resolve the poblem of over-trained XXXXX', ['neural network', 'parameter modification'])\n", "('NONE', 'algorithms used in the study are XXXXX ( es ) , XXXXX ( ga ) , differential evolution ( de ) , adaptive differential evolution ( jade ) , particle swarm optimization ( pso ) , artificial bee colony ( abc ) , cuckoo search ( cs ) and differential search algorithm ( ds )', ['evolution strategy', 'genetic algorithm'])\n", "('NONE', 'algorithms used in the study are XXXXX ( es ) , genetic algorithm ( ga ) , differential evolution ( de ) , adaptive differential evolution ( jade ) , XXXXX ( pso ) , artificial bee colony ( abc ) , cuckoo search ( cs ) and differential search algorithm ( ds )', ['evolution strategy', 'particle swarm optimization'])\n", "('method used for task', 'algorithms used in the study are XXXXX ( es ) , genetic algorithm ( ga ) , differential evolution ( de ) , adaptive differential evolution ( jade ) , particle swarm optimization ( pso ) , artificial bee colony ( abc ) , XXXXX ( cs ) and differential search algorithm ( ds )', ['evolution strategy', 'cuckoo search'])\n", "('method used for task', 'algorithms used in the study are XXXXX ( es ) , genetic algorithm ( ga ) , differential evolution ( de ) , adaptive differential evolution ( jade ) , particle swarm optimization ( pso ) , artificial bee colony ( abc ) , cuckoo search ( cs ) and differential XXXXX ( ds )', ['evolution strategy', 'search algorithm'])\n", "('NONE', 'algorithms used in the study are evolution strategy ( es ) , XXXXX ( ga ) , differential evolution ( de ) , adaptive differential evolution ( jade ) , XXXXX ( pso ) , artificial bee colony ( abc ) , cuckoo search ( cs ) and differential search algorithm ( ds )', ['genetic algorithm', 'particle swarm optimization'])\n", "('method used for task', 'algorithms used in the study are evolution strategy ( es ) , XXXXX ( ga ) , differential evolution ( de ) , adaptive differential evolution ( jade ) , particle swarm optimization ( pso ) , artificial bee colony ( abc ) , XXXXX ( cs ) and differential search algorithm ( ds )', ['genetic algorithm', 'cuckoo search'])\n", "('method used for task', 'algorithms used in the study are evolution strategy ( es ) , XXXXX ( ga ) , differential evolution ( de ) , adaptive differential evolution ( jade ) , particle swarm optimization ( pso ) , artificial bee colony ( abc ) , cuckoo search ( cs ) and differential XXXXX ( ds )', ['genetic algorithm', 'search algorithm'])\n", "('NONE', 'algorithms used in the study are evolution strategy ( es ) , genetic algorithm ( ga ) , differential evolution ( de ) , adaptive differential evolution ( jade ) , XXXXX ( pso ) , artificial bee colony ( abc ) , XXXXX ( cs ) and differential search algorithm ( ds )', ['particle swarm optimization', 'cuckoo search'])\n", "('method used for task', 'algorithms used in the study are evolution strategy ( es ) , genetic algorithm ( ga ) , differential evolution ( de ) , adaptive differential evolution ( jade ) , XXXXX ( pso ) , artificial bee colony ( abc ) , cuckoo search ( cs ) and differential XXXXX ( ds )', ['particle swarm optimization', 'search algorithm'])\n", "('method used for task', 'algorithms used in the study are evolution strategy ( es ) , genetic algorithm ( ga ) , differential evolution ( de ) , adaptive differential evolution ( jade ) , particle swarm optimization ( pso ) , artificial bee colony ( abc ) , XXXXX ( cs ) and differential XXXXX ( ds )', ['cuckoo search', 'search algorithm'])\n", "('method used for task', 'the artificial bee colony algorithm and differential XXXXX are the most XXXXX', ['search algorithm', 'robust algorithms'])\n", "('method used for task', 'the artificial bee XXXXX and differential XXXXX are the most robust algorithms', ['search algorithm', 'colony algorithm'])\n", "('method used for task', 'the artificial bee XXXXX and differential search algorithm are the most XXXXX', ['robust algorithms', 'colony algorithm'])\n", "('method used for task', 'comprehensive features ( cf ) include XXXXX ( sa ) , technical analysis ( ta ) and trend-based XXXXX ( tbsm )', ['sentiment analysis', 'segmentation method'])\n", "('method used for task', 'XXXXX of each dimension is divided into many equal subregions . according to XXXXX in particles’ historical best position , some dominant subregions can be determined', ['search space', 'statistics information'])\n", "('method used for task', 'XXXXX for XXXXX is considered', ['fuzzy demand', 'location problem'])\n", "('NONE', 'the efficiency and the effectiveness of the XXXXX in examined using XXXXX', ['hybrid algorithm', 'numerical examples'])\n", "('NONE', 'XXXXX are one of the most common choices reported in the literature for tuning XXXXX', ['evolutionary algorithms', 'fuzzy logic controllers'])\n", "('method used for task', 'an alternative is the simple XXXXX , which is a methodology designed to improve the response of type-1 XXXXX', ['fuzzy logic controllers', 'tuning algorithm'])\n", "('method used for task', 'an extension of the simple XXXXX for type-2 XXXXX is presented', ['fuzzy controllers', 'tuning algorithm'])\n", "('NONE', 'to incorporate XXXXX a novel XXXXX is defined', ['contextual information', 'energy function'])\n", "('NONE', 'the XXXXX of XXXXX for jusras problem grows exponentially with the number of users and receives antennas', ['computational complexity', 'exhaustive search'])\n", "('method used for task', 'we apply XXXXX ( bpso ) to the joint XXXXX and receive antenna selection problem', ['binary particle swarm optimization', 'user scheduling'])\n", "('NONE', 'in addition to applying the conventional bpso to jusras , we also present a specific improvement to this population-based XXXXX ; namely , we feed cyclically shifted XXXXX , so that the number of iterations until reaching an acceptable solution is reduced', ['heuristic algorithm', 'initial population'])\n", "('method used for task', 'a multi-agent based XXXXX is proposed for XXXXX', ['optimization approach', 'single machine scheduling problem'])\n", "('method used for task', 'an approach based on evolutionary and XXXXX is proposed for solving the XXXXX in crop growth dynamic models', ['bio-inspired algorithms', 'parameter estimation problem'])\n", "('NONE', 'XXXXX showed the best performance in solving the XXXXX of a dynamic crop growth model', ['differential evolution algorithm', 'parameter estimation problem'])\n", "('method used for task', 'a XXXXX and anova were applied to quantitatively evaluate the efficiency and effectiveness of XXXXX , cma-es , pso , abc and lse algorithms', ['statistical analysis', 'differential evolution'])\n", "('method used for task', 'this work presents flmicmac , an improved version of micmac based on computing with words , XXXXX and XXXXX', ['fuzzy numbers', 'linguistic variables'])\n", "('NONE', 'a new method is proposed for the exact analytical XXXXX of decomposable ts XXXXX with singleton and linear consequents', ['fuzzy systems', 'inverse mapping'])\n", "('NONE', 'the proposed method simplifies the inversion of decomposable ts XXXXX having high number of XXXXX', ['fuzzy systems', 'input variables'])\n", "('method used for task', 'we discuss a general framework of XXXXX for XXXXX and time series analysis', ['computational intelligence', 'signal processing'])\n", "('method used for task', 'we discuss a general framework of XXXXX for signal processing and XXXXX', ['computational intelligence', 'time series analysis'])\n", "('method used for task', 'we discuss a general framework of computational intelligence for XXXXX and XXXXX', ['signal processing', 'time series analysis'])\n", "('method used for task', 'multi-criteria decision making model is established for XXXXX and XXXXX', ['supplier evaluation', 'selection problem'])\n", "('method used for task', 'XXXXX ( nn ) for fuzzy multi criteria decision making and adaptive neuro-fuzzy XXXXX ( anfis ) models are compared for multi-criteria decision making in supplier evaluation and selection problem solution', ['neural networks', 'inference system'])\n", "('method used for task', 'XXXXX ( nn ) for fuzzy multi criteria decision making and adaptive neuro-fuzzy inference system ( anfis ) models are compared for XXXXX in supplier evaluation and selection problem solution', ['neural networks', 'multi-criteria decision making'])\n", "('method used for task', 'XXXXX ( nn ) for fuzzy multi criteria decision making and adaptive neuro-fuzzy inference system ( anfis ) models are compared for multi-criteria decision making in XXXXX and selection problem solution', ['neural networks', 'supplier evaluation'])\n", "('method used for task', 'neural networks ( nn ) for fuzzy multi criteria decision making and adaptive neuro-fuzzy XXXXX ( anfis ) models are compared for XXXXX in supplier evaluation and selection problem solution', ['inference system', 'multi-criteria decision making'])\n", "('method used for task', 'neural networks ( nn ) for fuzzy multi criteria decision making and adaptive neuro-fuzzy XXXXX ( anfis ) models are compared for multi-criteria decision making in XXXXX and selection problem solution', ['inference system', 'supplier evaluation'])\n", "('NONE', 'neural networks ( nn ) for fuzzy multi criteria decision making and adaptive neuro-fuzzy inference system ( anfis ) models are compared for XXXXX in XXXXX and selection problem solution', ['multi-criteria decision making', 'supplier evaluation'])\n", "('NONE', 'XXXXX of the XXXXX have found with ahp', ['performance criteria', 'importance weights'])\n", "('method used for task', 'the proposed approach is a combination of XXXXX ( pso ) and XXXXX ( lfpso )', ['particle swarm optimization', 'levy flight'])\n", "('NONE', 'the performance and accuracy of the lfpso are examined on numerical XXXXX especially XXXXX', ['benchmark functions', 'multimodal functions'])\n", "('method used for task', 'in addition , to evaluate achievement of the proposed method , the lfpso algorithm is compared the other XXXXX and other methods . the lfpso outperforms the XXXXX and other algorithms and it is closely successful with the XXXXX', ['abc algorithm', 'pso variants'])\n", "('method used for task', 'in addition , to evaluate achievement of the proposed method , the lfpso algorithm is compared the other XXXXX and other methods . the lfpso outperforms the XXXXX and other algorithms and it is closely successful with the XXXXX', ['abc algorithm', 'pso variants'])\n", "('method used for task', 'grnn is proposed to modify XXXXX and XXXXX decomposed pitch residuals', ['vocal tract', 'wavelet packet'])\n", "('NONE', 'XXXXX of grnn reduces XXXXX and overtraining of conventional ann', ['fast convergence', 'computation time'])\n", "('method used for task', 'we propose two novel evolutionary XXXXX approaches ( seann and teann ) for XXXXX ( tsf )', ['neural network', 'time series forecasting'])\n", "('NONE', 'for the weighted aggregation method , give a search method for finding XXXXX located in concave region on the XXXXX', ['pareto-optimal solutions', 'pareto front'])\n", "('method used for task', 'for the weighted aggregation method , give a XXXXX for finding XXXXX located in concave region on the pareto front', ['pareto-optimal solutions', 'search method'])\n", "('method used for task', 'for the weighted aggregation method , give a XXXXX for finding pareto-optimal solutions located in concave region on the XXXXX', ['pareto front', 'search method'])\n", "('method used for task', 'a hybridization of the XXXXX with nelder–mead method to produce an effective and efficient XXXXX', ['global optimization', 'search algorithm'])\n", "('method used for task', 'a set of XXXXX on synthetic signals shows the good performance with respect to fuzzy piecewise linear XXXXX', ['computational experiments', 'regression models'])\n", "('method used for task', 'a combined XXXXX and gradient-based optimization algorithm is applied for automatic selection of proper input variables and the model-dependent variable , and optimizing the XXXXX simultaneously', ['genetic optimization', 'model parameters'])\n", "('method used for task', 'a combined XXXXX and gradient-based optimization algorithm is applied for automatic selection of proper XXXXX and the model-dependent variable , and optimizing the model parameters simultaneously', ['genetic optimization', 'input variables'])\n", "('method used for task', 'a combined genetic optimization and gradient-based optimization algorithm is applied for automatic selection of proper XXXXX and the model-dependent variable , and optimizing the XXXXX simultaneously', ['model parameters', 'input variables'])\n", "('method used for task', 'XXXXX is applied to underwater glider XXXXX', ['differential evolution algorithm', 'path planning'])\n", "('method used for task', 'XXXXX for XXXXX', ['group decision making', 'medical diagnosis'])\n", "('NONE', 'XXXXX of fuzzy uncertain beam equation using XXXXX subject to unit step and impulse loads', ['numerical solution', 'adomian decomposition method'])\n", "('NONE', 'fuzziness appeared in the XXXXX are modeled through convex normalized XXXXX viz . triangular fuzzy numbers', ['initial conditions', 'fuzzy sets'])\n", "('NONE', 'fuzziness appeared in the XXXXX are modeled through convex normalized fuzzy sets viz . XXXXX', ['initial conditions', 'triangular fuzzy numbers'])\n", "('NONE', 'fuzziness appeared in the initial conditions are modeled through convex normalized XXXXX viz . XXXXX', ['fuzzy sets', 'triangular fuzzy numbers'])\n", "('method used for task', 'XXXXX ( adm ) and double parametric form is used with fuzzy based approach to obtain the uncertain bounds of the XXXXX', ['adomian decomposition method', 'dynamic responses'])\n", "('method used for task', 'classification using XXXXX is 99 % or more with near zero XXXXX', ['random forests', 'error measures'])\n", "('NONE', 'the present study analyzes the XXXXX of a repairable industrial system utilizing XXXXX', ['fuzzy reliability', 'uncertain data'])\n", "('method used for task', 'the problem is that when XXXXX is too large , XXXXX can not be used', ['state space', 'model checking'])\n", "('method used for task', 'we suggest using XXXXX to avoid searching the entire XXXXX', ['genetic algorithm', 'state space'])\n", "('NONE', 'the approach involves XXXXX or particle swarm XXXXX', ['simulated annealing', 'optimization algorithms'])\n", "('method used for task', 'we propose a novel hybrid XXXXX ( hvns ) algorithm for XXXXX ( hfs ) scheduling problems', ['variable neighborhood search', 'hybrid flow shop'])\n", "('method used for task', 'we propose a novel hybrid XXXXX ( hvns ) algorithm for hybrid flow shop ( hfs ) XXXXX', ['variable neighborhood search', 'scheduling problems'])\n", "('NONE', 'we propose a novel hybrid variable neighborhood search ( hvns ) algorithm for XXXXX ( hfs ) XXXXX', ['hybrid flow shop', 'scheduling problems'])\n", "('NONE', 'considering the XXXXX , eight XXXXX are developed', ['neighborhood structures', 'problem structure'])\n", "('method used for task', 'address the potential of XXXXX to forecast ground-level ozone in XXXXX', ['learning machine', 'urban area'])\n", "('method used for task', 'XXXXX of parameters for igsa is discussed on XXXXX', ['sensitivity analysis', 'test functions'])\n", "('method used for task', 'the overall approach of dnsa is to update the XXXXX dynamically and repair its XXXXX', ['graph model', 'spanning tree'])\n", "('NONE', 'XXXXX where gray coding is applied , improves the performance of the algorithm in XXXXX', ['parallel processing', 'large-scale problems'])\n", "('method used for task', 'XXXXX are employed to tuning the XXXXX', ['optimization algorithms', 'controller parameters'])\n", "('method used for task', 'a XXXXX is proposed to control a nonlinear XXXXX', ['hybrid method', 'dynamic system'])\n", "('NONE', 'this XXXXX combines XXXXX and kohonen algorithm to obtain faster convergence', ['hybrid algorithm', 'gradient method'])\n", "('method used for task', 'a XXXXX is designed by combining the ga , rbf-nn and XXXXX approaches', ['hybrid model', 'sugeno fuzzy logic'])\n", "('NONE', 'the rbf-nn is used to enhance the pid parameters obtained from ga to design sugeno XXXXX tuned by XXXXX ( ke , τe )', ['fuzzy pid controller', 'excitation parameter'])\n", "('method used for task', 'XXXXX based on XXXXX is designed to solve the proposed model', ['fuzzy simulation', 'tabu search algorithm'])\n", "('NONE', 'this paper proposes a new approach for training XXXXX with a XXXXX determination system', ['support vector machines', 'bone age'])\n", "('method used for task', 'the proposed approach is a combination of XXXXX ( pso ) and XXXXX ( svms )', ['particle swarm optimization', 'support vector machines'])\n", "('NONE', 'for comparison purposes , 6 nonlinear regression , 6 XXXXX , 5 anfis , 3 online XXXXX denfis , 3 offline XXXXX denfis , and 3 offline high-order tsk denfis models were developed', ['neural network', 'first-order tsk'])\n", "('NONE', 'for comparison purposes , 6 nonlinear regression , 6 XXXXX , 5 anfis , 3 online XXXXX denfis , 3 offline XXXXX denfis , and 3 offline high-order tsk denfis models were developed', ['neural network', 'first-order tsk'])\n", "('NONE', 'for comparison purposes , 6 nonlinear regression , 6 neural network , 5 anfis , 3 online XXXXX denfis , 3 offline XXXXX denfis , and 3 offline high-order tsk denfis models were developed', ['first-order tsk', 'first-order denfis'])\n", "('NONE', 'for comparison purposes , 6 nonlinear regression , 6 neural network , 5 anfis , 3 online XXXXX denfis , 3 offline XXXXX denfis , and 3 offline high-order tsk denfis models were developed', ['first-order tsk', 'first-order denfis'])\n", "('NONE', 'for comparison purposes , 6 nonlinear regression , 6 neural network , 5 anfis , 3 online XXXXX denfis , 3 offline XXXXX denfis , and 3 offline high-order tsk denfis models were developed', ['first-order tsk', 'high-order denfis'])\n", "('NONE', 'for comparison purposes , 6 nonlinear regression , 6 neural network , 5 anfis , 3 online XXXXX denfis , 3 offline XXXXX denfis , and 3 offline high-order tsk denfis models were developed', ['first-order denfis', 'first-order tsk'])\n", "('NONE', 'for comparison purposes , 6 nonlinear regression , 6 neural network , 5 anfis , 3 online XXXXX denfis , 3 offline XXXXX denfis , and 3 offline high-order tsk denfis models were developed', ['first-order tsk', 'first-order denfis'])\n", "('NONE', 'for comparison purposes , 6 nonlinear regression , 6 neural network , 5 anfis , 3 online XXXXX denfis , 3 offline XXXXX denfis , and 3 offline high-order tsk denfis models were developed', ['first-order tsk', 'high-order denfis'])\n", "('method used for task', 'XXXXX model could be trained to produce more reliable prediction results in comparison with XXXXX , anfis and nonlinear regression models', ['neural network', 'high-order denfis'])\n", "('NONE', 'a new fractal dimensional XXXXX on XXXXX is proposed', ['model based', 'fuzzy sets theory'])\n", "('NONE', 'we propose a tlbo algorithm for XXXXX of x-bar XXXXX', ['economic design', 'control chart'])\n", "('method used for task', 'we propose a XXXXX for XXXXX of x-bar control chart', ['economic design', 'tlbo algorithm'])\n", "('NONE', 'we propose a XXXXX for economic design of x-bar XXXXX', ['control chart', 'tlbo algorithm'])\n", "('method used for task', 'results of XXXXX are helpful to quality engineers in identifying the significant cost and XXXXX', ['sensitivity analysis', 'process parameters'])\n", "('NONE', 'tlbo is a promising method for the XXXXX of x-bar XXXXX', ['economic design', 'control chart'])\n", "('method used for task', 'a learning-guided XXXXX for constrained XXXXX is proposed', ['multi-objective evolutionary algorithm', 'portfolio optimization problem'])\n", "('method used for task', 'an extended version of weighted aggregated XXXXX assessment ( waspas method ) is proposed for XXXXX', ['soft computing', 'sum product'])\n", "('NONE', 'this study proposes a new method for XXXXX utilizing pso combined with a XXXXX', ['gene selection', 'decision tree'])\n", "('NONE', 'the approach is able to solve XXXXX when XXXXX on the number of clusters is not available', ['data clustering', 'prior knowledge'])\n", "('NONE', 'we have employed XXXXX , XXXXX , k-nearest neighbors , and support vector machine to improve churn prediction', ['decision tree', 'artificial neural networks'])\n", "('NONE', 'we have employed XXXXX , artificial neural networks , XXXXX , and support vector machine to improve churn prediction', ['decision tree', 'k-nearest neighbors'])\n", "('method used for task', 'we have employed XXXXX , artificial neural networks , k-nearest neighbors , and XXXXX to improve churn prediction', ['decision tree', 'support vector machine'])\n", "('method used for task', 'we have employed XXXXX , artificial neural networks , k-nearest neighbors , and support vector machine to improve XXXXX', ['decision tree', 'churn prediction'])\n", "('NONE', 'we have employed decision tree , XXXXX , XXXXX , and support vector machine to improve churn prediction', ['artificial neural networks', 'k-nearest neighbors'])\n", "('method used for task', 'we have employed decision tree , XXXXX , k-nearest neighbors , and XXXXX to improve churn prediction', ['artificial neural networks', 'support vector machine'])\n", "('method used for task', 'we have employed decision tree , XXXXX , k-nearest neighbors , and support vector machine to improve XXXXX', ['artificial neural networks', 'churn prediction'])\n", "('method used for task', 'we have employed decision tree , artificial neural networks , XXXXX , and XXXXX to improve churn prediction', ['k-nearest neighbors', 'support vector machine'])\n", "('method used for task', 'we have employed decision tree , artificial neural networks , XXXXX , and support vector machine to improve XXXXX', ['k-nearest neighbors', 'churn prediction'])\n", "('method used for task', 'we have employed decision tree , artificial neural networks , k-nearest neighbors , and XXXXX to improve XXXXX', ['support vector machine', 'churn prediction'])\n", "('NONE', 'the aim of this paper is to solve the problem of decision makings by introducing soft XXXXX in XXXXX', ['discernibility matrix', 'soft sets'])\n", "('method used for task', 'the notion of soft XXXXX is firstly introduced in XXXXX', ['discernibility matrix', 'soft sets'])\n", "('method used for task', 'an novel algorithm based on the soft XXXXX is proposed to solve the problems of XXXXX', ['discernibility matrix', 'decision making'])\n", "('method used for task', 'the weighted soft XXXXX is introduced in XXXXX and its application to decision making is also investigated', ['discernibility matrix', 'soft sets'])\n", "('method used for task', 'the weighted soft XXXXX is introduced in soft sets and its application to XXXXX is also investigated', ['discernibility matrix', 'decision making'])\n", "('method used for task', 'the weighted soft discernibility matrix is introduced in XXXXX and its application to XXXXX is also investigated', ['soft sets', 'decision making'])\n", "('NONE', 'converting the XXXXX into XXXXX matrices', ['stability problem', 'system parameter'])\n", "('NONE', 'XXXXX within an interval-valued XXXXX', ['multiple criteria decision analysis', 'fuzzy environment'])\n", "('method used for task', 'a XXXXX is implemented with computational XXXXX', ['comparative study', 'experimental analysis'])\n", "('method used for task', 'presenting self-tuning XXXXX based on XXXXX emotional learning to control dvr', ['pi controller', 'human brain'])\n", "('method used for task', 'XXXXX and final answer are great better in tlbo algorithm than XXXXX', ['convergence speed', 'pso algorithm'])\n", "('method used for task', 'XXXXX and final answer are great better in XXXXX than pso algorithm', ['convergence speed', 'tlbo algorithm'])\n", "('NONE', 'convergence speed and final answer are great better in XXXXX than XXXXX', ['pso algorithm', 'tlbo algorithm'])\n", "('NONE', 'the proposed XXXXX enhances the achieved strip thickness under high XXXXX', ['interval type-2 fuzzy logic system', 'uncertainty level'])\n", "('NONE', 'we take covering XXXXX as models of e-spread XXXXX', ['approximation spaces', 'information systems'])\n", "('method used for task', 'we solved the XXXXX for different cases and different XXXXX', ['optimal power flow', 'test systems'])\n", "('NONE', 'bhbo is conceptually very simple , further unlike other XXXXX bhbo parameter-less XXXXX', ['optimization techniques', 'optimization technique'])\n", "('method used for task', 'XXXXX for XXXXX', ['wireless sensor networks', 'soft computing techniques'])\n", "('NONE', 'we utilize XXXXX ( XXXXX and classification ) on ct images', ['image analysis', 'feature extraction'])\n", "('method used for task', 'we utilize XXXXX ( feature extraction and classification ) on XXXXX', ['image analysis', 'ct images'])\n", "('method used for task', 'we utilize image analysis ( XXXXX and classification ) on XXXXX', ['feature extraction', 'ct images'])\n", "('method used for task', 'an evolutionary XXXXX was implemented to develop a XXXXX on a comprehensive field measurements database', ['model based', 'data mining technique'])\n", "('method used for task', 'neutrosophic XXXXX is defined to describe the XXXXX on images', ['similarity function', 'uncertain information'])\n", "('method used for task', 'a novel XXXXX is defined using neutrosophic XXXXX and the new defined clustering algorithm classifies the pixels on the image into different groups', ['objective function', 'similarity function'])\n", "('method used for task', 'the proposed decision framework is able to derive XXXXX for building cluster which could significantly reduce XXXXX', ['pareto solutions', 'energy cost'])\n", "('method used for task', 'the proposed XXXXX is able to derive XXXXX for building cluster which could significantly reduce energy cost', ['pareto solutions', 'decision framework'])\n", "('method used for task', 'the proposed XXXXX is able to derive pareto solutions for building cluster which could significantly reduce XXXXX', ['energy cost', 'decision framework'])\n", "('NONE', 'compare to a XXXXX based decision framework , the proposed decision framework could significantly reduce XXXXX', ['memetic algorithm', 'computational cost'])\n", "('NONE', 'compare to a XXXXX based XXXXX , the proposed XXXXX could significantly reduce computational cost', ['memetic algorithm', 'decision framework'])\n", "('NONE', 'compare to a XXXXX based XXXXX , the proposed XXXXX could significantly reduce computational cost', ['memetic algorithm', 'decision framework'])\n", "('NONE', 'compare to a memetic algorithm based XXXXX , the proposed XXXXX could significantly reduce XXXXX', ['computational cost', 'decision framework'])\n", "('NONE', 'compare to a memetic algorithm based XXXXX , the proposed XXXXX could significantly reduce XXXXX', ['computational cost', 'decision framework'])\n", "('method used for task', 'we introduce a novel XXXXX to finding the XXXXX of a nonlinear function', ['iterative method', 'fixed point'])\n", "('method used for task', 'we combine ideas proposed in artificial bee XXXXX and XXXXX', ['bisection method', 'colony algorithm'])\n", "('NONE', 'it considers another uncertainty in the XXXXX of XXXXX', ['membership function', 'fuzzy set'])\n", "('NONE', 'the XXXXX is one of the leading XXXXX', ['multiobjective evolutionary algorithms', 'moea/d algorithm'])\n", "('NONE', 'XXXXX is a branch of XXXXX in which the agent must learn through interaction with the environment', ['reinforcement learning', 'machine learning'])\n", "('method used for task', 'XXXXX and XXXXX are conducted as a unifying model selection', ['feature selection', 'parameter tuning'])\n", "('method used for task', 'XXXXX and parameter tuning are conducted as a unifying XXXXX', ['feature selection', 'model selection'])\n", "('NONE', 'feature selection and XXXXX are conducted as a unifying XXXXX', ['parameter tuning', 'model selection'])\n", "('method used for task', 'we present a study about the performance of two classical weight training methods of XXXXX ( rbfn ) , least mean square ( lms ) and XXXXX ( svd ) , applied to classification problems , when the data-sets are imbalanced', ['radial basis function networks', 'singular value decomposition'])\n", "('method used for task', 'we present a study about the performance of two classical weight training methods of XXXXX ( rbfn ) , least mean square ( lms ) and singular value decomposition ( svd ) , applied to XXXXX , when the data-sets are imbalanced', ['radial basis function networks', 'classification problems'])\n", "('method used for task', 'we present a study about the performance of two classical weight training methods of radial basis function networks ( rbfn ) , least mean square ( lms ) and XXXXX ( svd ) , applied to XXXXX , when the data-sets are imbalanced', ['singular value decomposition', 'classification problems'])\n", "('NONE', 'XXXXX permitted the analysis of the behavior of our algorithm for different categories of XXXXX', ['experimental evaluation', 'mobile applications'])\n", "('method used for task', 'we present a review paper on the importance of data granulation in the wide contexts of XXXXX and XXXXX', ['soft computing', 'pattern recognition'])\n", "('NONE', 'we present a XXXXX on the importance of data granulation in the wide contexts of XXXXX and pattern recognition', ['soft computing', 'review paper'])\n", "('NONE', 'we present a XXXXX on the importance of data granulation in the wide contexts of soft computing and XXXXX', ['pattern recognition', 'review paper'])\n", "('NONE', 'we extend the topsis approach to rank all the alternatives from the perspective of the magnitude of XXXXX under interval-valued intuitionistic XXXXX', ['fuzzy environment', 'decision information'])\n", "('NONE', 'analytic XXXXX ( ahp ) was proposed to achieve weights of criteria in a XXXXX', ['hierarchy process', 'group decision-making problem'])\n", "('NONE', 'the XXXXX could be provided by the user or could be the result of performing a conventional XXXXX over the input data', ['clustering method', 'label information'])\n", "('NONE', 'we develop XXXXX ( ann , fis , and anfis ) for estimating the number of XXXXX based on the occurrence of economic improvement projects', ['adverse events', 'soft computing techniques'])\n", "('NONE', 'we develop soft computing techniques ( ann , fis , and anfis ) for estimating the number of XXXXX based on the occurrence of economic XXXXX', ['adverse events', 'improvement projects'])\n", "('NONE', 'we develop XXXXX ( ann , fis , and anfis ) for estimating the number of adverse events based on the occurrence of economic XXXXX', ['soft computing techniques', 'improvement projects'])\n", "('NONE', 'when the XXXXX was calculated based on the mape for each of the models , ann had better XXXXX than fis and anfis models , as demonstrated by experimental results', ['model accuracy', 'predictive accuracy'])\n", "('NONE', 'in this paper we introduce an adapted elitist XXXXX for automatic knot adjustment of XXXXX', ['clonal selection algorithm', 'b-spline curves'])\n", "('method used for task', 'adaptive XXXXX ( agfs ) for optimizing rules and XXXXX', ['genetic fuzzy system', 'membership functions'])\n", "('NONE', 'we address the problem of XXXXX in multidimensional XXXXX', ['missing data', 'time series'])\n", "('method used for task', 'de novo XXXXX supplies novel molecules for XXXXX', ['drug design', 'drug development'])\n", "('method used for task', 'XXXXX are used for XXXXX', ['evolutionary algorithms', 'multi-objective optimization'])\n", "('NONE', 'we create XXXXX ( fis ) as a means of computerizing XXXXX ( dd ) tables', ['fuzzy inference systems', 'differential diagnosis'])\n", "('method used for task', 'XXXXX based on a simulator are performed to validate the effectiveness of the proposed XXXXX', ['numerical experiments', 'decision rules'])\n", "('NONE', 'since the key space of vigenere cryptosystem is very large , therefore , brute force does not help in the analysis of the cryptosystem in XXXXX . XXXXX are also not helpful when the key size is not small', ['real time', 'statistical techniques'])\n", "('NONE', 'the aim of this work was to examine the feasibility of XXXXX as a cryptanalysis tool of classical cryptosystems especially vigenere cipher . we applied XXXXX , particle swarm optimization and cuckoo search techniques for the analysis of vigenere cipher . after doing lots of experiments we observed that ga and pso techniques can recover the entire key of vigenere cipher correctly for keys of small lengths . however , cuckoo search can recover more than 90 % of the key characters for keys of size up to 25 characters because ga and pso get trapped in local minima while cuckoo search finds the global optimum solution following characteristics of lévy flights', ['cuckoo search algorithm', 'genetic algorithm'])\n", "('NONE', 'the aim of this work was to examine the feasibility of XXXXX as a cryptanalysis tool of classical cryptosystems especially vigenere cipher . we applied genetic algorithm , XXXXX and cuckoo search techniques for the analysis of vigenere cipher . after doing lots of experiments we observed that ga and pso techniques can recover the entire key of vigenere cipher correctly for keys of small lengths . however , cuckoo search can recover more than 90 % of the key characters for keys of size up to 25 characters because ga and pso get trapped in local minima while cuckoo search finds the global optimum solution following characteristics of lévy flights', ['cuckoo search algorithm', 'particle swarm optimization'])\n", "('NONE', 'the aim of this work was to examine the feasibility of XXXXX as a cryptanalysis tool of classical cryptosystems especially vigenere cipher . we applied genetic algorithm , particle swarm optimization and XXXXX techniques for the analysis of vigenere cipher . after doing lots of experiments we observed that ga and pso techniques can recover the entire key of vigenere cipher correctly for keys of small lengths . however , XXXXX can recover more than 90 % of the key characters for keys of size up to 25 characters because ga and pso get trapped in local minima while XXXXX finds the global optimum solution following characteristics of lévy flights', ['cuckoo search algorithm', 'cuckoo search'])\n", "('NONE', 'the aim of this work was to examine the feasibility of XXXXX as a cryptanalysis tool of classical cryptosystems especially vigenere cipher . we applied genetic algorithm , particle swarm optimization and cuckoo search techniques for the analysis of vigenere cipher . after doing lots of experiments we observed that ga and pso techniques can recover the entire key of vigenere cipher correctly for keys of small lengths . however , cuckoo search can recover more than 90 % of the key characters for keys of size up to 25 characters because ga and pso get trapped in XXXXX while cuckoo search finds the global optimum solution following characteristics of lévy flights', ['cuckoo search algorithm', 'local minima'])\n", "('NONE', 'the aim of this work was to examine the feasibility of XXXXX as a cryptanalysis tool of classical cryptosystems especially vigenere cipher . we applied genetic algorithm , particle swarm optimization and XXXXX techniques for the analysis of vigenere cipher . after doing lots of experiments we observed that ga and pso techniques can recover the entire key of vigenere cipher correctly for keys of small lengths . however , XXXXX can recover more than 90 % of the key characters for keys of size up to 25 characters because ga and pso get trapped in local minima while XXXXX finds the global optimum solution following characteristics of lévy flights', ['cuckoo search algorithm', 'cuckoo search'])\n", "('NONE', 'the aim of this work was to examine the feasibility of cuckoo search algorithm as a cryptanalysis tool of classical cryptosystems especially vigenere cipher . we applied XXXXX , XXXXX and cuckoo search techniques for the analysis of vigenere cipher . after doing lots of experiments we observed that ga and pso techniques can recover the entire key of vigenere cipher correctly for keys of small lengths . however , cuckoo search can recover more than 90 % of the key characters for keys of size up to 25 characters because ga and pso get trapped in local minima while cuckoo search finds the global optimum solution following characteristics of lévy flights', ['genetic algorithm', 'particle swarm optimization'])\n", "('NONE', 'the aim of this work was to examine the feasibility of XXXXX algorithm as a cryptanalysis tool of classical cryptosystems especially vigenere cipher . we applied XXXXX , particle swarm optimization and XXXXX techniques for the analysis of vigenere cipher . after doing lots of experiments we observed that ga and pso techniques can recover the entire key of vigenere cipher correctly for keys of small lengths . however , XXXXX can recover more than 90 % of the key characters for keys of size up to 25 characters because ga and pso get trapped in local minima while XXXXX finds the global optimum solution following characteristics of lévy flights', ['genetic algorithm', 'cuckoo search'])\n", "('NONE', 'the aim of this work was to examine the feasibility of cuckoo search algorithm as a cryptanalysis tool of classical cryptosystems especially vigenere cipher . we applied XXXXX , particle swarm optimization and cuckoo search techniques for the analysis of vigenere cipher . after doing lots of experiments we observed that ga and pso techniques can recover the entire key of vigenere cipher correctly for keys of small lengths . however , cuckoo search can recover more than 90 % of the key characters for keys of size up to 25 characters because ga and pso get trapped in XXXXX while cuckoo search finds the global optimum solution following characteristics of lévy flights', ['genetic algorithm', 'local minima'])\n", "('NONE', 'the aim of this work was to examine the feasibility of XXXXX algorithm as a cryptanalysis tool of classical cryptosystems especially vigenere cipher . we applied XXXXX , particle swarm optimization and XXXXX techniques for the analysis of vigenere cipher . after doing lots of experiments we observed that ga and pso techniques can recover the entire key of vigenere cipher correctly for keys of small lengths . however , XXXXX can recover more than 90 % of the key characters for keys of size up to 25 characters because ga and pso get trapped in local minima while XXXXX finds the global optimum solution following characteristics of lévy flights', ['genetic algorithm', 'cuckoo search'])\n", "('NONE', 'the aim of this work was to examine the feasibility of XXXXX algorithm as a cryptanalysis tool of classical cryptosystems especially vigenere cipher . we applied genetic algorithm , XXXXX and XXXXX techniques for the analysis of vigenere cipher . after doing lots of experiments we observed that ga and pso techniques can recover the entire key of vigenere cipher correctly for keys of small lengths . however , XXXXX can recover more than 90 % of the key characters for keys of size up to 25 characters because ga and pso get trapped in local minima while XXXXX finds the global optimum solution following characteristics of lévy flights', ['particle swarm optimization', 'cuckoo search'])\n", "('NONE', 'the aim of this work was to examine the feasibility of cuckoo search algorithm as a cryptanalysis tool of classical cryptosystems especially vigenere cipher . we applied genetic algorithm , XXXXX and cuckoo search techniques for the analysis of vigenere cipher . after doing lots of experiments we observed that ga and pso techniques can recover the entire key of vigenere cipher correctly for keys of small lengths . however , cuckoo search can recover more than 90 % of the key characters for keys of size up to 25 characters because ga and pso get trapped in XXXXX while cuckoo search finds the global optimum solution following characteristics of lévy flights', ['particle swarm optimization', 'local minima'])\n", "('NONE', 'the aim of this work was to examine the feasibility of XXXXX algorithm as a cryptanalysis tool of classical cryptosystems especially vigenere cipher . we applied genetic algorithm , XXXXX and XXXXX techniques for the analysis of vigenere cipher . after doing lots of experiments we observed that ga and pso techniques can recover the entire key of vigenere cipher correctly for keys of small lengths . however , XXXXX can recover more than 90 % of the key characters for keys of size up to 25 characters because ga and pso get trapped in local minima while XXXXX finds the global optimum solution following characteristics of lévy flights', ['particle swarm optimization', 'cuckoo search'])\n", "('NONE', 'the aim of this work was to examine the feasibility of XXXXX algorithm as a cryptanalysis tool of classical cryptosystems especially vigenere cipher . we applied genetic algorithm , particle swarm optimization and XXXXX techniques for the analysis of vigenere cipher . after doing lots of experiments we observed that ga and pso techniques can recover the entire key of vigenere cipher correctly for keys of small lengths . however , XXXXX can recover more than 90 % of the key characters for keys of size up to 25 characters because ga and pso get trapped in XXXXX while XXXXX finds the global optimum solution following characteristics of lévy flights', ['cuckoo search', 'local minima'])\n", "('NONE', 'the aim of this work was to examine the feasibility of XXXXX algorithm as a cryptanalysis tool of classical cryptosystems especially vigenere cipher . we applied genetic algorithm , particle swarm optimization and XXXXX techniques for the analysis of vigenere cipher . after doing lots of experiments we observed that ga and pso techniques can recover the entire key of vigenere cipher correctly for keys of small lengths . however , XXXXX can recover more than 90 % of the key characters for keys of size up to 25 characters because ga and pso get trapped in XXXXX while XXXXX finds the global optimum solution following characteristics of lévy flights', ['local minima', 'cuckoo search'])\n", "('NONE', \"idsfq is presented to provide adaptive control scenario for dynamic XXXXX inside cloud 's XXXXX\", ['resource provisioning', 'spot market'])\n", "('NONE', 'a new XXXXX of the protein folding problem in 2d-hp model is proposed for the use of XXXXX', ['state space representation', 'reinforcement learning methods'])\n", "('NONE', 'the proposed XXXXX also provides an actual learning for an agent . thus , at the end of a XXXXX an agent could find the optimum fold of any sequence of a certain length , which is not the case in the existing reinforcement learning methods', ['state space representation', 'learning process'])\n", "('NONE', 'the proposed XXXXX also provides an actual learning for an agent . thus , at the end of a learning process an agent could find the optimum fold of any sequence of a certain length , which is not the case in the existing XXXXX', ['state space representation', 'reinforcement learning methods'])\n", "('NONE', 'the proposed state space representation also provides an actual learning for an agent . thus , at the end of a XXXXX an agent could find the optimum fold of any sequence of a certain length , which is not the case in the existing XXXXX', ['learning process', 'reinforcement learning methods'])\n", "('method used for task', 'by using the ant-q algorithm ( an ant based reinforcement learning method ) , optimum fold of a XXXXX is found rapidly when compared to the standard XXXXX', ['protein sequence', 'q-learning algorithm'])\n", "('method used for task', 'a simpler approach to XXXXX based on XXXXX is presented with covering based rough sets', ['attribute reduction', 'discernibility matrix'])\n", "('method used for task', 'a simpler approach to XXXXX based on discernibility matrix is presented with covering based XXXXX', ['attribute reduction', 'rough sets'])\n", "('method used for task', 'a XXXXX to XXXXX based on discernibility matrix is presented with covering based rough sets', ['attribute reduction', 'simpler approach'])\n", "('method used for task', 'a simpler approach to attribute reduction based on XXXXX is presented with covering based XXXXX', ['discernibility matrix', 'rough sets'])\n", "('method used for task', 'a XXXXX to attribute reduction based on XXXXX is presented with covering based rough sets', ['discernibility matrix', 'simpler approach'])\n", "('method used for task', 'a XXXXX to attribute reduction based on discernibility matrix is presented with covering based XXXXX', ['rough sets', 'simpler approach'])\n", "('NONE', 'some important properties of XXXXX with covering based XXXXX are improved', ['attribute reduction', 'rough sets'])\n", "('NONE', 'a new algorithm to XXXXX in XXXXX is presented in a different strategy of identifying objects', ['attribute reduction', 'decision tables'])\n", "('NONE', 'design of three unsupervised XXXXX that satisfying exactly XXXXX', ['initial conditions', 'ann models'])\n", "('method used for task', 'a XXXXX based on modified servqual model and XXXXX is used', ['hybrid method', 'fuzzy set theory'])\n", "('method used for task', 'improvement in performance by asset utilization , XXXXX and XXXXX in terms of production loss', ['energy efficient', 'cost reduction'])\n", "('method used for task', 'a case based reasoning system with an original XXXXX for biomedical XXXXX is proposed', ['hidden markov model', 'text classification'])\n", "('NONE', 'the model of XXXXX with XXXXX and the seasonal mechanism is proposed', ['support vector regression', 'adaptive genetic algorithm'])\n", "('method used for task', 'XXXXX ( rl ) is applied to construct the XXXXX', ['reinforcement learning', 'probabilistic model'])\n", "('method used for task', 'this paper demonstrates that the recently discovered XXXXX is a better technique for the optimization of large XXXXX', ['firefly algorithm', 'wind farms'])\n", "('NONE', 'it compares the XXXXX with the XXXXX or by the use of spread sheet methods such as finite difference method', ['firefly algorithm', 'genetic algorithms'])\n", "('NONE', 'it compares the XXXXX with the genetic algorithms or by the use of spread sheet methods such as XXXXX', ['firefly algorithm', 'finite difference method'])\n", "('NONE', 'it compares the firefly algorithm with the XXXXX or by the use of spread sheet methods such as XXXXX', ['genetic algorithms', 'finite difference method'])\n", "('NONE', 'this paper demonstrates that the XXXXX outperforms the XXXXX and finite difference method by a wide margin', ['firefly algorithm', 'genetic algorithms'])\n", "('NONE', 'this paper demonstrates that the XXXXX outperforms the genetic algorithms and XXXXX by a wide margin', ['firefly algorithm', 'finite difference method'])\n", "('method used for task', 'this paper demonstrates that the firefly algorithm outperforms the XXXXX and XXXXX by a wide margin', ['genetic algorithms', 'finite difference method'])\n", "('NONE', 'the crack growth rate has been calculated by using an XXXXX from experimental crack length ( a ) and number of cycles ( n ) data which has been subsequently used as training data base for gp XXXXX along with load ratio ( r ) , maximum stress intensity factor ( kmax ) and stress intensity factor rage ( δk )', ['exponential model', 'model formulation'])\n", "('NONE', 'the crack growth rate has been calculated by using an XXXXX from experimental crack length ( a ) and number of cycles ( n ) data which has been subsequently used as training data base for gp model formulation along with load ratio ( r ) , maximum XXXXX ( kmax ) and XXXXX rage ( δk )', ['exponential model', 'stress intensity factor'])\n", "('NONE', 'the crack growth rate has been calculated by using an XXXXX from experimental crack length ( a ) and number of cycles ( n ) data which has been subsequently used as training data base for gp model formulation along with load ratio ( r ) , maximum XXXXX ( kmax ) and XXXXX rage ( δk )', ['exponential model', 'stress intensity factor'])\n", "('NONE', 'the XXXXX has been calculated by using an XXXXX from experimental crack length ( a ) and number of cycles ( n ) data which has been subsequently used as training data base for gp model formulation along with load ratio ( r ) , maximum stress intensity factor ( kmax ) and stress intensity factor rage ( δk )', ['exponential model', 'crack growth rate'])\n", "('NONE', 'the crack growth rate has been calculated by using an exponential model from experimental crack length ( a ) and number of cycles ( n ) data which has been subsequently used as training data base for gp XXXXX along with load ratio ( r ) , maximum XXXXX ( kmax ) and XXXXX rage ( δk )', ['model formulation', 'stress intensity factor'])\n", "('NONE', 'the crack growth rate has been calculated by using an exponential model from experimental crack length ( a ) and number of cycles ( n ) data which has been subsequently used as training data base for gp XXXXX along with load ratio ( r ) , maximum XXXXX ( kmax ) and XXXXX rage ( δk )', ['model formulation', 'stress intensity factor'])\n", "('NONE', 'the XXXXX has been calculated by using an exponential model from experimental crack length ( a ) and number of cycles ( n ) data which has been subsequently used as training data base for gp XXXXX along with load ratio ( r ) , maximum stress intensity factor ( kmax ) and stress intensity factor rage ( δk )', ['model formulation', 'crack growth rate'])\n", "('NONE', 'the XXXXX has been calculated by using an exponential model from experimental crack length ( a ) and number of cycles ( n ) data which has been subsequently used as training data base for gp model formulation along with load ratio ( r ) , maximum XXXXX ( kmax ) and XXXXX rage ( δk )', ['stress intensity factor', 'crack growth rate'])\n", "('NONE', 'the XXXXX has been calculated by using an exponential model from experimental crack length ( a ) and number of cycles ( n ) data which has been subsequently used as training data base for gp model formulation along with load ratio ( r ) , maximum XXXXX ( kmax ) and XXXXX rage ( δk )', ['stress intensity factor', 'crack growth rate'])\n", "('NONE', 'the validity of the proposed gp model has been confirmed by comparing the XXXXX by XXXXX and also with previously proposed ann model', ['experimental data', 'model prediction'])\n", "('NONE', 'the validity of the proposed gp model has been confirmed by comparing the model prediction by XXXXX and also with previously proposed XXXXX', ['experimental data', 'ann model'])\n", "('NONE', 'the validity of the proposed gp model has been confirmed by comparing the XXXXX by experimental data and also with previously proposed XXXXX', ['model prediction', 'ann model'])\n", "('NONE', 'nowadays , software designers attempt to employ XXXXX in software design phase , but XXXXX may be not used in the implementation phase . therefore , one of the challenging issues is XXXXX of source code and design , i.e. , XXXXX', ['design patterns', 'conformance checking'])\n", "('NONE', 'nowadays , software designers attempt to employ XXXXX in software design phase , but XXXXX may be not used in the implementation phase . therefore , one of the challenging issues is conformance checking of XXXXX and design , i.e. , XXXXX', ['design patterns', 'source code'])\n", "('NONE', 'nowadays , software designers attempt to employ XXXXX in software design phase , but XXXXX may be not used in the implementation phase . therefore , one of the challenging issues is XXXXX of source code and design , i.e. , XXXXX', ['design patterns', 'conformance checking'])\n", "('NONE', 'nowadays , software designers attempt to employ XXXXX in software design phase , but XXXXX may be not used in the implementation phase . therefore , one of the challenging issues is conformance checking of XXXXX and design , i.e. , XXXXX', ['design patterns', 'source code'])\n", "('NONE', 'nowadays , software designers attempt to employ design patterns in software design phase , but design patterns may be not used in the implementation phase . therefore , one of the challenging issues is XXXXX of XXXXX and design , i.e. , design patterns', ['conformance checking', 'source code'])\n", "('NONE', 'nowadays , software designers attempt to employ XXXXX in software design phase , but XXXXX may be not used in the implementation phase . therefore , one of the challenging issues is XXXXX of source code and design , i.e. , XXXXX', ['conformance checking', 'design patterns'])\n", "('NONE', 'nowadays , software designers attempt to employ XXXXX in software design phase , but XXXXX may be not used in the implementation phase . therefore , one of the challenging issues is conformance checking of XXXXX and design , i.e. , XXXXX', ['source code', 'design patterns'])\n", "('NONE', 'in addition , after developing a system , usually its documents are not maintained , so , identifying XXXXX from XXXXX can help to achieve the design of an existing system as a reverse engineering task', ['design pattern', 'source code'])\n", "('NONE', 'the variant implementations ( i.e. , different XXXXX ) of a design pattern make hard to detect the design pattern instances from the XXXXX . to address this issue , in this paper , we propose a new method which aims to map the design pattern detection problem into a learning problem', ['source code', 'source codes'])\n", "('method used for task', 'the variant implementations ( i.e. , different source codes ) of a design pattern make hard to detect the design pattern instances from the XXXXX . to address this issue , in this paper , we propose a new method which aims to map the design pattern XXXXX into a learning problem', ['source code', 'detection problem'])\n", "('method used for task', 'the variant implementations ( i.e. , different XXXXX ) of a design pattern make hard to detect the design pattern instances from the source code . to address this issue , in this paper , we propose a new method which aims to map the design pattern XXXXX into a learning problem', ['source codes', 'detection problem'])\n", "('method used for task', 'central XXXXX ( cfo ) is deterministic XXXXX', ['metaheuristic algorithm', 'force optimization'])\n", "('NONE', 'XXXXX are applied for predicting the development effort of XXXXX', ['neural networks', 'software projects'])\n", "('NONE', 'XXXXX of XXXXX is compared with that of a statistical regression', ['prediction accuracy', 'neural networks'])\n", "('NONE', 'a XXXXX had a XXXXX with the highest confidence level', ['radial basis function neural network', 'prediction accuracy'])\n", "('NONE', 'a XXXXX had a prediction accuracy with the highest XXXXX', ['radial basis function neural network', 'confidence level'])\n", "('NONE', 'a radial basis function neural network had a XXXXX with the highest XXXXX', ['prediction accuracy', 'confidence level'])\n", "('method used for task', 'the XXXXX was based on a genetic algorithm-based XXXXX', ['neural fuzzy system', 'software sensor'])\n", "('method used for task', 'fuzzy XXXXX and XXXXX was used to identify the model', ['subtractive clustering', 'genetic algorithm'])\n", "('method used for task', 'we introduce the concept of XXXXX in binary spaces.it is proven that utilizing random numbers and their opposite is beneficial in evolutionary algorithms.opposite numbers are applied to accelerate the XXXXX of binary gravitational search algorithm ( bgsa ) .the results show that obgsa possesses superior performance in accuracy as compared to the bgsa', ['opposition-based learning', 'convergence rate'])\n", "('method used for task', 'we introduce the concept of XXXXX in binary spaces.it is proven that utilizing random numbers and their opposite is beneficial in evolutionary algorithms.opposite numbers are applied to accelerate the convergence rate of binary XXXXX ( bgsa ) .the results show that obgsa possesses superior performance in accuracy as compared to the bgsa', ['opposition-based learning', 'gravitational search algorithm'])\n", "('NONE', 'we introduce the concept of opposition-based learning in binary spaces.it is proven that utilizing random numbers and their opposite is beneficial in evolutionary algorithms.opposite numbers are applied to accelerate the XXXXX of binary XXXXX ( bgsa ) .the results show that obgsa possesses superior performance in accuracy as compared to the bgsa', ['convergence rate', 'gravitational search algorithm'])\n", "('method used for task', 'we reviewed the use of XXXXX on the mdvrp ( multi depot XXXXX )', ['genetic algorithms', 'vehicle routing problem'])\n", "('method used for task', 'we compared the XXXXX to other XXXXX on mdvrp based on the results on standard benchmarks', ['genetic algorithms', 'metaheuristic algorithms'])\n", "('method used for task', 'XXXXX ( pso ) is used to optimize the XXXXX ( svm )', ['particle swarm optimization', 'support vector machine'])\n", "('method used for task', 'the gm ( 1 , n ) model and XXXXX are used for XXXXX , which is capable of simplifying the system prediction model', ['fuzzy c-means clustering', 'variable selection'])\n", "('method used for task', 'proposing acor–pso as a two-stage meta-heuristic XXXXX to efficiently find the XXXXX of double-arch concrete dams', ['optimization model', 'optimal shape'])\n", "('NONE', 'reducing the XXXXX of optimization procedure of double-arch dams subject to earthquake loads using XXXXX', ['computational time', 'weighted least squares support vector machine'])\n", "('method used for task', 'this paper proposes a new method for speaker XXXXX based on formants , wavelet entropy and XXXXX denoted as fwenn', ['feature extraction', 'neural networks'])\n", "('method used for task', 'mapping the XXXXX according to XXXXX', ['image histogram', 'rayleigh distribution'])\n", "('NONE', 'the proposed flow-based XXXXX represents a complete departure from the traditional XXXXX rather than modifies simply the traditional XXXXX', ['similarity measure', 'distance measure'])\n", "('NONE', 'the proposed flow-based XXXXX represents a complete departure from the traditional XXXXX rather than modifies simply the traditional XXXXX', ['similarity measure', 'distance measure'])\n", "('NONE', 'XXXXX performs the following : summarize XXXXX for sfp models', ['systematic literature review', 'ml techniques'])\n", "('NONE', 'assess XXXXX and capability of XXXXX for constructing sfp models', ['performance accuracy', 'ml techniques'])\n", "('NONE', 'provide comparison of XXXXX of different XXXXX', ['performance accuracy', 'ml techniques'])\n", "('NONE', 'we review applications of the XXXXX in the statistical XXXXX', ['time series analysis', 'soft computing techniques'])\n", "('method used for task', 'the employed XXXXX and XXXXX provide useful information for forecasting', ['data mining', 'classification methods'])\n", "('NONE', 'to realize the XXXXX of implementing XXXXX planning system', ['risk factor', 'enterprise resource'])\n", "('NONE', 'a new XXXXX based hybrid XXXXX is proposed to further improve the performance of the algorithm', ['neighborhood structure', 'mutation strategy'])\n", "('method used for task', 'XXXXX is performed under wide changes in XXXXX and disturbance', ['robustness analysis', 'system parameters'])\n", "('NONE', 'based on the introduced XXXXX , two types of XXXXX are proposed whose consequent parts of the rules are linear and quadratic functions', ['membership function', 'ts models'])\n", "('method used for task', 'an XXXXX is suggested to identify the proposed XXXXX', ['incremental learning algorithm', 'ts models'])\n", "('NONE', 'obtained results demonstrate XXXXX and low redundancy of the suggested XXXXX', ['high accuracy', 'ts fuzzy models'])\n", "('method used for task', 'nonconvex emission constrained XXXXX ( neced ) is a complex XXXXX', ['economic dispatch', 'optimization problem'])\n", "('NONE', 'the framework can be used in XXXXX , like XXXXX', ['engineering applications', 'video surveillance'])\n", "('method used for task', 'combination of XXXXX and entropy method is applied for XXXXX weighting', ['fuzzy ahp', 'risk factor'])\n", "('method used for task', 'combination of XXXXX and XXXXX is applied for risk factor weighting', ['fuzzy ahp', 'entropy method'])\n", "('method used for task', 'combination of fuzzy ahp and XXXXX is applied for XXXXX weighting', ['risk factor', 'entropy method'])\n", "('method used for task', 'fuzzy XXXXX is used to determine the risk priorities of XXXXX', ['vikor method', 'failure modes'])\n", "('method used for task', 'this paper also ventures on a new problem on XXXXX compared to previous smaller scale data used in XXXXX of efg in ref . [ 2 ] and text extraction task in refs . [ 1,6 ]', ['text classification', 'case study'])\n", "('NONE', 'we use an improved multiobjective estimation of distribution algorithm ( irm-meda ) to solve the environmental XXXXX of hydrothermal XXXXX', ['economic dispatch', 'power systems'])\n", "('method used for task', 'we use an improved multiobjective estimation of XXXXX ( irm-meda ) to solve the environmental XXXXX of hydrothermal power systems', ['economic dispatch', 'distribution algorithm'])\n", "('method used for task', 'we use an improved multiobjective estimation of XXXXX ( irm-meda ) to solve the environmental economic dispatch of hydrothermal XXXXX', ['power systems', 'distribution algorithm'])\n", "('method used for task', 'the proposed algorithm obtained new optimal solutions and XXXXX for XXXXX', ['upper bounds', 'benchmark problems'])\n", "('NONE', 'a new method of XXXXX of the generalized XXXXX', ['similarity measure', 'trapezoidal fuzzy numbers'])\n", "('method used for task', 'combining ga and tmvdr achieve XXXXX and XXXXX', ['high accuracy', 'fast convergence'])\n", "('NONE', 'to our knowledge , it was the first study in identifying the XXXXX with XXXXX', ['computer vision', 'butterfly species'])\n", "('method used for task', 'the method is based on XXXXX and XXXXX', ['local binary patterns', 'artificial neural network'])\n", "('method used for task', \"a novel XXXXX based on improved culture algorithm for a vehicle 's active suspension was proposed and supported by XXXXX\", ['control strategy', 'numerical simulation'])\n", "('method used for task', \"a novel XXXXX based on improved XXXXX for a vehicle 's active suspension was proposed and supported by numerical simulation\", ['control strategy', 'culture algorithm'])\n", "('method used for task', \"a novel control strategy based on improved XXXXX for a vehicle 's active suspension was proposed and supported by XXXXX\", ['numerical simulation', 'culture algorithm'])\n", "('NONE', 'by simulation , the perfect control effect is obtained by using fuzzy XXXXX with the optimal XXXXX optimized by improved culture algorithm', ['pid control', 'fuzzy rules'])\n", "('NONE', 'by simulation , the perfect control effect is obtained by using fuzzy XXXXX with the optimal fuzzy rules optimized by improved XXXXX', ['pid control', 'culture algorithm'])\n", "('NONE', 'by simulation , the perfect control effect is obtained by using fuzzy pid control with the optimal XXXXX optimized by improved XXXXX', ['fuzzy rules', 'culture algorithm'])\n", "('method used for task', 'proposing a novel framework for evaluating teaching performance based on the combination of XXXXX and fuzzy comprehensive XXXXX', ['fuzzy ahp', 'evaluation method'])\n", "('method used for task', 'the output gain of the adaptive XXXXX ( afc ) , it is considerate as a XXXXX', ['fuzzy controller', 'fuzzy variable'])\n", "('NONE', 'the comparative at XXXXX study , shows that the adaptive XXXXX can be enhance the performances the dc voltage control at various operating', ['pi controller', 'fuzzy controller'])\n", "('method used for task', 'XXXXX is developed by combining XXXXX ( ga ) , radial basis function neural network ( rbf-nn ) and sugeno fuzzy logic approaches', ['integrated model', 'genetic algorithm'])\n", "('method used for task', 'XXXXX is developed by combining genetic algorithm ( ga ) , XXXXX ( rbf-nn ) and sugeno fuzzy logic approaches', ['integrated model', 'radial basis function neural network'])\n", "('method used for task', 'XXXXX is developed by combining genetic algorithm ( ga ) , radial basis function neural network ( rbf-nn ) and XXXXX approaches', ['integrated model', 'sugeno fuzzy logic'])\n", "('NONE', 'integrated model is developed by combining XXXXX ( ga ) , XXXXX ( rbf-nn ) and sugeno fuzzy logic approaches', ['genetic algorithm', 'radial basis function neural network'])\n", "('method used for task', 'integrated model is developed by combining XXXXX ( ga ) , radial basis function neural network ( rbf-nn ) and XXXXX approaches', ['genetic algorithm', 'sugeno fuzzy logic'])\n", "('method used for task', 'integrated model is developed by combining genetic algorithm ( ga ) , XXXXX ( rbf-nn ) and XXXXX approaches', ['radial basis function neural network', 'sugeno fuzzy logic'])\n", "('NONE', \"enhanced pid parameters are used to design sugeno XXXXX tuned by XXXXX ( ke , τe ) of the avr system to improve the system 's response\", ['fuzzy pid controller', 'excitation parameter'])\n", "('NONE', \"enhanced pid parameters are used to design sugeno XXXXX tuned by excitation parameter ( ke , τe ) of the XXXXX to improve the system 's response\", ['fuzzy pid controller', 'avr system'])\n", "('NONE', \"enhanced pid parameters are used to design sugeno fuzzy pid controller tuned by XXXXX ( ke , τe ) of the XXXXX to improve the system 's response\", ['excitation parameter', 'avr system'])\n", "('method used for task', 'we proposed a new multiobjective XXXXX based methodology for adaption of neural network structure for XXXXX of satellite imagery', ['particle swarm optimization', 'pixel classification'])\n", "('method used for task', 'we define a intuitionistic fuzzy parameterized XXXXX for dealing with uncertainties that is based on both XXXXX and XXXXX', ['soft sets', 'intuitionistic fuzzy sets'])\n", "('method used for task', 'we define a intuitionistic fuzzy parameterized XXXXX for dealing with uncertainties that is based on both XXXXX and XXXXX', ['soft sets', 'intuitionistic fuzzy sets'])\n", "('method used for task', 'the fuzziness of the XXXXX is analyzed in two different ways such as – triangular and XXXXX', ['trapezoidal fuzzy numbers', 'credit period'])\n", "('NONE', 'reformulation of the involved XXXXX in terms of penalized XXXXX for the pso implementation', ['constrained optimization problem', 'unconstrained optimization problem'])\n", "('method used for task', 'we have experimentally validated the optimization capabilities of these schemes and compared them to XXXXX , dynamic mesh optimization , and XXXXX', ['particle swarm optimization', 'firefly algorithm'])\n", "('NONE', 'an XXXXX that segments and classifies four XXXXX', ['integrated system', 'moving objects'])\n", "('method used for task', 'automatic XXXXX for aann and XXXXX and extraction methods are developed', ['data generation', 'data selection'])\n", "('method used for task', 'automatic XXXXX for aann and data selection and XXXXX are developed', ['data generation', 'extraction methods'])\n", "('method used for task', 'automatic data generation for aann and XXXXX and XXXXX are developed', ['data selection', 'extraction methods'])\n", "('method used for task', 'an estimation model of XXXXX based on XXXXX', ['missing data', 'multilayer perceptron'])\n", "('NONE', 'an XXXXX of XXXXX based on multilayer perceptron', ['missing data', 'estimation model'])\n", "('NONE', 'an XXXXX of missing data based on XXXXX', ['multilayer perceptron', 'estimation model'])\n", "('method used for task', 'combination of XXXXX and k-nearest neighbour-based XXXXX', ['neural network', 'multiple imputation'])\n", "('NONE', 'the optimization of the antecedent parameters for a type 2 XXXXX of XXXXX is presented', ['fuzzy system', 'edge detection'])\n", "('NONE', 'the goal of XXXXX in XXXXX is to provide the ability to handle uncertainty', ['interval type-2 fuzzy logic', 'edge detection methods'])\n", "('NONE', 'results show that the XXXXX provides better results in optimizing the type-2 XXXXX', ['cuckoo search', 'fuzzy system'])\n", "('method used for task', 'the main disadvantage of fwnn is that the application domain is limited to static problems due to its feed-forward XXXXX . therefore , we propose to use a self-recurrent XXXXX ( srwnn ) in the consequent part of fwnn , solving the control problem for chaotic systems', ['network structure', 'wavelet neural network'])\n", "('method used for task', 'the main disadvantage of fwnn is that the application domain is limited to static problems due to its feed-forward XXXXX . therefore , we propose to use a self-recurrent wavelet neural network ( srwnn ) in the consequent part of fwnn , solving the control problem for XXXXX', ['network structure', 'chaotic systems'])\n", "('NONE', 'the main disadvantage of fwnn is that the application domain is limited to static problems due to its feed-forward network structure . therefore , we propose to use a self-recurrent XXXXX ( srwnn ) in the consequent part of fwnn , solving the control problem for XXXXX', ['wavelet neural network', 'chaotic systems'])\n", "('NONE', 'finding the optimal learning rates is a challenging task in the classic gradient-based XXXXX . hence , in our proposed framework , all of the learning rates are determined optimally based on XXXXX', ['learning algorithms', 'lyapunov stability theory'])\n", "('method used for task', 'finding the optimal XXXXX is a challenging task in the classic gradient-based XXXXX . hence , in our proposed framework , all of the XXXXX are determined optimally based on lyapunov stability theory', ['learning algorithms', 'learning rates'])\n", "('method used for task', 'finding the optimal XXXXX is a challenging task in the classic gradient-based XXXXX . hence , in our proposed framework , all of the XXXXX are determined optimally based on lyapunov stability theory', ['learning algorithms', 'learning rates'])\n", "('NONE', 'finding the optimal XXXXX is a challenging task in the classic gradient-based learning algorithms . hence , in our proposed framework , all of the XXXXX are determined optimally based on XXXXX', ['lyapunov stability theory', 'learning rates'])\n", "('NONE', 'finding the optimal XXXXX is a challenging task in the classic gradient-based learning algorithms . hence , in our proposed framework , all of the XXXXX are determined optimally based on XXXXX', ['lyapunov stability theory', 'learning rates'])\n", "('method used for task', 'the merits of proposed methodology are XXXXX and less XXXXX', ['high accuracy', 'execution time'])\n", "('NONE', 'having the best performance of np-pso in solving various XXXXX compared with some well-known XXXXX', ['nonlinear functions', 'pso algorithms'])\n", "('NONE', 'XXXXX ( rbf ) XXXXX is developed on fpga', ['radial basis function', 'neural network'])\n", "('method used for task', 'XXXXX on randomly generated graphs and XXXXX of results', ['computational experiments', 'statistical analysis'])\n", "('NONE', 'the joint statistics and XXXXX of the pdtdfb XXXXX are studied', ['mutual information', 'transform coefficients'])\n", "('NONE', 'the pdtdfb transform coefficients are modeled using a hmt XXXXX with XXXXX', ['statistical model', 'gaussian mixtures'])\n", "('NONE', 'the pdtdfb XXXXX are modeled using a hmt XXXXX with gaussian mixtures', ['statistical model', 'transform coefficients'])\n", "('NONE', 'the pdtdfb XXXXX are modeled using a hmt statistical model with XXXXX', ['gaussian mixtures', 'transform coefficients'])\n", "('NONE', 'interaction of XXXXX with sbst combines XXXXX with experience', ['computational power', 'domain experts'])\n", "('NONE', 'the length of XXXXX of XXXXX could be adjusted adaptively to balance the computing time and solving ability', ['scatter search', 'reference set'])\n", "('method used for task', 'the length of reference set of XXXXX could be adjusted adaptively to balance the XXXXX and solving ability', ['scatter search', 'computing time'])\n", "('NONE', 'the length of XXXXX of scatter search could be adjusted adaptively to balance the XXXXX and solving ability', ['reference set', 'computing time'])\n", "('method used for task', 'based on karnik-mendel ( km ) algorithm , we propose an XXXXX to XXXXX , and apply it in personnel selection problems for knowledge-intensive enterprise', ['analytical solution', 'fuzzy topsis method'])\n", "('method used for task', 'orpd problem is formulated for real XXXXX and XXXXX minimization', ['power loss', 'voltage deviation'])\n", "('method used for task', 'soft computing techniques used : ontologies , XXXXX , and a XXXXX', ['statistical tests', 'classification algorithm'])\n", "('NONE', 'XXXXX used : ontologies , XXXXX , and a classification algorithm', ['statistical tests', 'soft computing techniques'])\n", "('method used for task', 'XXXXX used : ontologies , statistical tests , and a XXXXX', ['classification algorithm', 'soft computing techniques'])\n", "('method used for task', 'the new XXXXX without insignificant components are established through the XXXXX', ['regression models', 'stepwise regression'])\n", "('method used for task', 'adaptively computes the XXXXX for XXXXX and restoration purposes', ['fuzzy membership functions', 'noise detection'])\n", "('method used for task', 'selection of XXXXX and controller structure is vital for XXXXX', ['objective function', 'controller design'])\n", "('method used for task', 'selection of XXXXX and XXXXX is vital for controller design', ['objective function', 'controller structure'])\n", "('method used for task', 'selection of objective function and XXXXX is vital for XXXXX', ['controller design', 'controller structure'])\n", "('NONE', 'an XXXXX using itae , XXXXX and settling times is proposed', ['objective function', 'damping ratio'])\n", "('method used for task', 'apply XXXXX to dynamic XXXXX', ['hopfield neural network', 'parameter estimation'])\n", "('NONE', 'obtain the initial XXXXX of regular XXXXX', ['fuzzy classification', 'feature space'])\n", "('NONE', 'the objective is to reach standard levels of XXXXX with applying XXXXX of passive filters', ['harmonic distortion', 'minimum cost'])\n", "('method used for task', 'publicly available XXXXX were used for a XXXXX', ['environmental data', 'comparative study'])\n", "('method used for task', 'seven important XXXXX found to exist in XXXXX 50–300hz . a new method of selecting cwt scales according to the XXXXX found', ['fault frequencies', 'frequency range'])\n", "('method used for task', 'seven important XXXXX found to exist in XXXXX 50–300hz . a new method of selecting cwt scales according to the XXXXX found', ['frequency range', 'fault frequencies'])\n", "('method used for task', 'the mofca aims to balance the XXXXX over network by wisely adjusting the cluster competition radius according to three XXXXX', ['energy consumption', 'input parameters'])\n", "('method used for task', 'the XXXXX is based on XXXXX', ['quality evaluation', 'fuzzy classification'])\n", "('method used for task', 'an XXXXX and a heuristic are presented to solve the XXXXX', ['integer programming', 'np-hard problem'])\n", "('NONE', 'proposing a five- tiers XXXXX with flexible XXXXX', ['supply chain', 'lead time'])\n", "('method used for task', 'XXXXX is indispensable when dealing with XXXXX', ['feature selection', 'microarray data'])\n", "('method used for task', 'combination of pnn , XXXXX and XXXXX', ['hierarchical clustering', 'cluster validity index'])\n", "('method used for task', 'multiobjective spatial XXXXX for XXXXX is proposed', ['fuzzy clustering', 'image segmentation'])\n", "('NONE', 'a premise synchronizer has been delicately constructed to ensure the same premises with uniform XXXXX in both XXXXX and fuzzy controller', ['time scales', 'fuzzy model'])\n", "('NONE', 'a premise synchronizer has been delicately constructed to ensure the same premises with uniform XXXXX in both fuzzy model and XXXXX', ['time scales', 'fuzzy controller'])\n", "('method used for task', 'a premise synchronizer has been delicately constructed to ensure the same premises with uniform time scales in both XXXXX and XXXXX', ['fuzzy model', 'fuzzy controller'])\n", "('method used for task', 'XXXXX are identified by fmea and XXXXX', ['critical parameters', 'fuzzy fmea'])\n", "('NONE', 'we propose an improved XXXXX which makes the best of ergodicity of piecewise linear XXXXX to explore the global search while utilizing the sequential quadratic programming to accelerate the local search', ['gravitational search algorithm', 'chaotic map'])\n", "('NONE', 'we propose an improved XXXXX which makes the best of ergodicity of piecewise linear chaotic map to explore the XXXXX while utilizing the sequential quadratic programming to accelerate the local search', ['gravitational search algorithm', 'global search'])\n", "('NONE', 'we propose an improved XXXXX which makes the best of ergodicity of piecewise linear chaotic map to explore the global search while utilizing the sequential quadratic programming to accelerate the XXXXX', ['gravitational search algorithm', 'local search'])\n", "('method used for task', 'we propose an improved gravitational search algorithm which makes the best of ergodicity of piecewise linear XXXXX to explore the XXXXX while utilizing the sequential quadratic programming to accelerate the local search', ['chaotic map', 'global search'])\n", "('method used for task', 'we propose an improved gravitational search algorithm which makes the best of ergodicity of piecewise linear XXXXX to explore the global search while utilizing the sequential quadratic programming to accelerate the XXXXX', ['chaotic map', 'local search'])\n", "('method used for task', 'we propose an improved gravitational search algorithm which makes the best of ergodicity of piecewise linear chaotic map to explore the XXXXX while utilizing the sequential quadratic programming to accelerate the XXXXX', ['global search', 'local search'])\n", "('method used for task', 'based on the binary improved XXXXX and the XXXXX ( k-nn ) method , a novel hybrid system is proposed to improve classification accuracy with an appropriate feature subset in binary problems', ['gravitational search algorithm', 'k-nearest neighbor'])\n", "('method used for task', 'based on the binary improved XXXXX and the k-nearest neighbor ( k-nn ) method , a novel XXXXX is proposed to improve classification accuracy with an appropriate feature subset in binary problems', ['gravitational search algorithm', 'hybrid system'])\n", "('method used for task', 'based on the binary improved XXXXX and the k-nearest neighbor ( k-nn ) method , a novel hybrid system is proposed to improve XXXXX with an appropriate feature subset in binary problems', ['gravitational search algorithm', 'classification accuracy'])\n", "('NONE', 'based on the binary improved gravitational search algorithm and the XXXXX ( k-nn ) method , a novel XXXXX is proposed to improve classification accuracy with an appropriate feature subset in binary problems', ['k-nearest neighbor', 'hybrid system'])\n", "('method used for task', 'based on the binary improved gravitational search algorithm and the XXXXX ( k-nn ) method , a novel hybrid system is proposed to improve XXXXX with an appropriate feature subset in binary problems', ['k-nearest neighbor', 'classification accuracy'])\n", "('method used for task', 'based on the binary improved gravitational search algorithm and the k-nearest neighbor ( k-nn ) method , a novel XXXXX is proposed to improve XXXXX with an appropriate feature subset in binary problems', ['hybrid system', 'classification accuracy'])\n", "('NONE', 'three XXXXX based XXXXX such as , ann–ga , ann–sa and ann–quasi newton have been developed', ['soft computing', 'integrated models'])\n", "('NONE', 'XXXXX on four different XXXXX ( non-neutral and neutral nk landscapes , flow-shop , qap , maxsat )', ['experimental analysis', 'landscape models'])\n", "('method used for task', 'm-band XXXXX is used to extract scale-space features for XXXXX', ['wavelet packet', 'document image'])\n", "('method used for task', 'er-wca outperforms or equals other methods in terms of XXXXX and XXXXX', ['function evaluations', 'solution quality'])\n", "('NONE', 'human visual perception in assessing XXXXX using type 2 XXXXX', ['image quality', 'fuzzy sets'])\n", "('method used for task', 'we design an XXXXX which is suitable for process with XXXXX', ['control chart', 'fuzzy parameters'])\n", "('NONE', 'we solve the XXXXX using the combination of two XXXXX', ['evolutionary methods', 'path planning problem'])\n", "('NONE', 'second , XXXXX ( ep ) optimizes the XXXXX and smoothness', ['evolutionary programming', 'path length'])\n", "('method used for task', 'also developed hybrid of XXXXX and adaptive XXXXX', ['score level', 'fuzzy decision level'])\n", "('NONE', 'the fmcdm model avoids multiplying two XXXXX into a pooled XXXXX , and reserves fuzzy information', ['fuzzy numbers', 'fuzzy number'])\n", "('NONE', 'the effect of degree of XXXXX and hybrid combination of XXXXX is studied', ['activation functions', 'input variables'])\n", "('method used for task', 'we propose a directional XXXXX for XXXXX ( de )', ['mutation operator', 'differential evolution'])\n", "('NONE', 'this XXXXX the permutation flowshop scheduling problem with XXXXX and weighted tardiness', ['order acceptance', 'paper studies'])\n", "('NONE', 'this paper studies the XXXXX with XXXXX and weighted tardiness', ['order acceptance', 'permutation flowshop scheduling problem'])\n", "('NONE', 'this XXXXX the XXXXX with order acceptance and weighted tardiness', ['paper studies', 'permutation flowshop scheduling problem'])\n", "('NONE', 'nearest XXXXX of continuous type-2 fuzzy variable with cv-based XXXXX', ['interval approximation', 'reduction method'])\n", "('NONE', 'a new XXXXX , combining XXXXX and adaptive-network-based fuzzy inference system ( anfis–pso ) was used', ['hybrid approach', 'particle swarm optimization'])\n", "('method used for task', 'a new XXXXX , combining particle swarm optimization and adaptive-network-based XXXXX ( anfis–pso ) was used', ['hybrid approach', 'fuzzy inference system'])\n", "('method used for task', 'a new hybrid approach , combining XXXXX and adaptive-network-based XXXXX ( anfis–pso ) was used', ['particle swarm optimization', 'fuzzy inference system'])\n", "('method used for task', 'a thresholding method is proposed using XXXXX and XXXXX', ['fuzzy entropy', 'bat algorithm'])\n", "('NONE', 'a XXXXX is proposed using XXXXX and bat algorithm', ['fuzzy entropy', 'thresholding method'])\n", "('method used for task', 'a XXXXX is proposed using fuzzy entropy and XXXXX', ['bat algorithm', 'thresholding method'])\n", "('NONE', 'the proposed method can stably converge to XXXXX with XXXXX', ['optimal threshold', 'high efficiency'])\n", "('method used for task', 'we present a bi-objective XXXXX for a XXXXX', ['inventory model', 'supply chain problem'])\n", "('method used for task', 'it2fls employs hybrid learning process by XXXXX and XXXXX', ['fuzzy c-means', 'genetic algorithm'])\n", "('NONE', 'to evaluate the gsd team-level service climate and gsd project outcome relationship based on adaptive neuro-fuzzy XXXXX ( anfis ) with the hybrid taguchi-genetic XXXXX ( htgla )', ['inference system', 'learning algorithm'])\n", "('method used for task', 'to evaluate the gsd team-level XXXXX and gsd project outcome relationship based on adaptive neuro-fuzzy XXXXX ( anfis ) with the hybrid taguchi-genetic learning algorithm ( htgla )', ['inference system', 'service climate'])\n", "('method used for task', 'to evaluate the gsd team-level XXXXX and gsd project outcome relationship based on adaptive neuro-fuzzy inference system ( anfis ) with the hybrid taguchi-genetic XXXXX ( htgla )', ['learning algorithm', 'service climate'])\n", "('NONE', 'our evolutionary XXXXX provides improved XXXXX', ['recognition rate', 'face recognition algorithm'])\n", "('NONE', 'we obtain the XXXXX of XXXXX induced by soft coverings', ['lattice structure', 'soft sets'])\n", "('method used for task', 'we investigate the parameter reduction of soft coverings by means of knowledge on the XXXXX for covering XXXXX', ['attribute reduction', 'information systems'])\n", "('NONE', 'we investigate the XXXXX of soft coverings by means of knowledge on the XXXXX for covering information systems', ['attribute reduction', 'parameter reduction'])\n", "('NONE', 'we investigate the XXXXX of soft coverings by means of knowledge on the attribute reduction for covering XXXXX', ['information systems', 'parameter reduction'])\n", "('NONE', 'the development of a XXXXX capable of determining the number of daily deals present on a given XXXXX , and accurately localizing those segments of the page that consist of information about one particular deal', ['segmentation algorithm', 'web page'])\n", "('method used for task', 'we propose XXXXX and XXXXX to solve the problem', ['genetic algorithm', 'particle swarm optimization'])\n", "('NONE', 'to improve XXXXX , aco based XXXXX for pmc is proposed', ['classification accuracy', 'feature selection'])\n", "('NONE', 'presenting an efficient XXXXX by combining XXXXX and relaxation induced neighborhood search to solve the proposed model', ['hybrid algorithm', 'local branching'])\n", "('method used for task', 'presenting an efficient XXXXX by combining local branching and relaxation induced XXXXX to solve the proposed model', ['hybrid algorithm', 'neighborhood search'])\n", "('method used for task', 'presenting an efficient hybrid algorithm by combining XXXXX and relaxation induced XXXXX to solve the proposed model', ['local branching', 'neighborhood search'])\n", "('NONE', 'the object is divided into m overlapped patches to improve XXXXX in XXXXX and partial occlusions', ['tracking performance', 'background clutter'])\n", "('NONE', 'during a XXXXX , a XXXXX is constructed to minimize the total probability of misclassifying the labeled training examples , a process which approximately maximizes the margins of the resulting classifier', ['decision tree', 'training phase'])\n", "('method used for task', 'we propose an adaptive takagi-sugeno-kang-fuzzy ( tsk-fuzzy ) XXXXX ( atfsc ) for use in direct torque control ( dtc ) induction motor ( im ) drives to improve their XXXXX', ['dynamic responses', 'speed controller'])\n", "('method used for task', 'we propose an adaptive takagi-sugeno-kang-fuzzy ( tsk-fuzzy ) speed controller ( atfsc ) for use in direct XXXXX ( dtc ) induction motor ( im ) drives to improve their XXXXX', ['dynamic responses', 'torque control'])\n", "('method used for task', 'we propose an adaptive takagi-sugeno-kang-fuzzy ( tsk-fuzzy ) XXXXX ( atfsc ) for use in direct XXXXX ( dtc ) induction motor ( im ) drives to improve their dynamic responses', ['speed controller', 'torque control'])\n", "('NONE', 'the atfsc , XXXXX , and pi control schemes were experimentally investigated , using the root mean square error ( rmse ) XXXXX to evaluate each scheme', ['fuzzy control', 'performance index'])\n", "('NONE', 'the atfsc , XXXXX , and pi control schemes were experimentally investigated , using the XXXXX square error ( rmse ) performance index to evaluate each scheme', ['fuzzy control', 'root mean'])\n", "('NONE', 'the atfsc , fuzzy control , and pi control schemes were experimentally investigated , using the XXXXX square error ( rmse ) XXXXX to evaluate each scheme', ['performance index', 'root mean'])\n", "('method used for task', 'the system uses : XXXXX and XXXXX to self-adjust the foa', ['genetic fuzzy systems', 'genetic programming'])\n", "('method used for task', 'this paper proposes an improved multi-objective XXXXX chemotaxis algorithm to solve XXXXX', ['bacteria colony', 'multi-objective optimization problems'])\n", "('method used for task', 'the XXXXX obtained by the improved algorithm has better distribution and convergence than other XXXXX', ['pareto front', 'optimization algorithms'])\n", "('NONE', 'the XXXXX of the general pareto-based multi-objective XXXXX chemotaxis algorithm is proved', ['convergence property', 'bacteria colony'])\n", "('method used for task', 'applications were done on some XXXXX and real-world datasets taken from uci XXXXX', ['benchmark data', 'machine learning repository'])\n", "('NONE', 'XXXXX are number of solar panels , XXXXX and batteries', ['wind turbines', 'decision variables'])\n", "('NONE', 'XXXXX ( de ) , XXXXX ( pso ) and game-theoretic XXXXX ( gtde ) algorithms are implemented', ['differential evolution', 'particle swarm optimization'])\n", "('NONE', 'XXXXX ( de ) , XXXXX ( pso ) and game-theoretic XXXXX ( gtde ) algorithms are implemented', ['particle swarm optimization', 'differential evolution'])\n", "('NONE', 'XXXXX provide adaptive XXXXX for public health protection', ['dynamic models', 'decision support'])\n", "('NONE', 'we presented new hardware architectures for XXXXX of several it2 flcs ( km , wm , bmm , nt ) with the consideration of limitations : device area and XXXXX', ['hardware implementation', 'real-time performance'])\n", "('method used for task', 'we presented new XXXXX for XXXXX of several it2 flcs ( km , wm , bmm , nt ) with the consideration of limitations : device area and real-time performance', ['hardware implementation', 'hardware architectures'])\n", "('NONE', 'we presented new XXXXX for hardware implementation of several it2 flcs ( km , wm , bmm , nt ) with the consideration of limitations : device area and XXXXX', ['real-time performance', 'hardware architectures'])\n", "('method used for task', 'we apply ga and XXXXX to find XXXXX for the entire network', ['pso algorithm', 'optimal solution'])\n", "('NONE', 'a bidirectional XXXXX r-variable XXXXX ( biorv-nsa ) is proposed to generate less mature detectors and cover more “black holes”', ['negative selection algorithm', 'inhibition optimization'])\n", "('NONE', 'when doing comparative experiments , biorv-nsa is divided into two kinds of algorithms , which are a bidirectional weak XXXXX r-variable XXXXX and a bidirectional strong XXXXX r-variable XXXXX , respectively', ['negative selection algorithm', 'inhibition optimization'])\n", "('NONE', 'when doing comparative experiments , biorv-nsa is divided into two kinds of algorithms , which are a bidirectional weak XXXXX r-variable XXXXX and a bidirectional strong XXXXX r-variable XXXXX , respectively', ['negative selection algorithm', 'inhibition optimization'])\n", "('NONE', 'when doing comparative experiments , biorv-nsa is divided into two kinds of algorithms , which are a bidirectional weak XXXXX r-variable XXXXX and a bidirectional strong XXXXX r-variable XXXXX , respectively', ['negative selection algorithm', 'inhibition optimization'])\n", "('NONE', 'when doing comparative experiments , biorv-nsa is divided into two kinds of algorithms , which are a bidirectional weak XXXXX r-variable XXXXX and a bidirectional strong XXXXX r-variable XXXXX , respectively', ['negative selection algorithm', 'inhibition optimization'])\n", "('method used for task', 'the objective of this novel XXXXX is to increase the XXXXX for data set with outliers', ['support vector regression', 'generalization ability'])\n", "('method used for task', 'a new XXXXX for power system disturbances is introduced using XXXXX', ['learning method', 'extreme learning machine'])\n", "('method used for task', 'simultaneously optimize the XXXXX and XXXXX for the elm using pso', ['model selection', 'feature subset'])\n", "('method used for task', 'proposed method can improve XXXXX and XXXXX of elm', ['generalization performance', 'convergence accuracy'])\n", "('NONE', 'XXXXX using XXXXX', ['principal component analysis', 'feature dimension reduction'])\n", "('NONE', 'XXXXX was built via XXXXX ( ann )', ['prediction model', 'artificial neural network'])\n", "('NONE', 'pe has better XXXXX in XXXXX', ['generalization ability', 'activity recognition'])\n", "('NONE', 'the study models the problem by bi-level XXXXX . this study develops a revised ant algorithm with efficient XXXXX to solve the problem', ['stochastic programming', 'greedy heuristics'])\n", "('method used for task', 'a XXXXX is proposed for efficient XXXXX in high-dimensional datasets . the symmetrical uncertainty ( su ) criterion is exploited to weight features in filter phase', ['hybrid method', 'subset selection'])\n", "('NONE', 'in wrapper phase , both fuzzy imperialist competitive algorithm ( fica ) and incremental wrapper XXXXX with replacement ( iwssr ) in weighted XXXXX are executed to search and find relevant attributes', ['subset selection', 'feature space'])\n", "('NONE', '31 components of XXXXX are tested with a systematic XXXXX', ['abc algorithm', 'experimental study'])\n", "('method used for task', 'two XXXXX , soco and cec05 , are used in XXXXX', ['experimental study', 'benchmark sets'])\n", "('NONE', 'two new variants of XXXXX are proposed for each of the two XXXXX', ['abc algorithms', 'benchmark sets'])\n", "('NONE', 'proposed methodology uses the current signal in XXXXX as the inputs of the pattern classifiers for XXXXX', ['time domain', 'fault diagnosis'])\n", "('NONE', 'we propose a hybrid collaborative XXXXX on fuzzy social-only XXXXX and local search methods for dynamic optimization problems', ['model based', 'particle swarm optimization'])\n", "('method used for task', 'we propose a hybrid collaborative XXXXX on fuzzy social-only particle swarm optimization and XXXXX for dynamic optimization problems', ['model based', 'local search methods'])\n", "('method used for task', 'we propose a hybrid collaborative model based on fuzzy social-only XXXXX and XXXXX for dynamic optimization problems', ['particle swarm optimization', 'local search methods'])\n", "('NONE', 'the XXXXX , in precise XXXXX ( rst ) based class information and edge information is used in bilateral framework for medical image denoising problem', ['soft computing', 'rough set theory'])\n", "('method used for task', 'the XXXXX , in precise rough set theory ( rst ) based class information and XXXXX is used in bilateral framework for medical image denoising problem', ['soft computing', 'edge information'])\n", "('NONE', 'the XXXXX , in precise rough set theory ( rst ) based XXXXX and edge information is used in bilateral framework for medical image denoising problem', ['soft computing', 'class information'])\n", "('method used for task', 'the soft computing , in precise XXXXX ( rst ) based class information and XXXXX is used in bilateral framework for medical image denoising problem', ['rough set theory', 'edge information'])\n", "('NONE', 'the soft computing , in precise XXXXX ( rst ) based XXXXX and edge information is used in bilateral framework for medical image denoising problem', ['rough set theory', 'class information'])\n", "('method used for task', 'the soft computing , in precise rough set theory ( rst ) based XXXXX and XXXXX is used in bilateral framework for medical image denoising problem', ['edge information', 'class information'])\n", "('method used for task', 'as the XXXXX , cart ( classification and XXXXX ) , lsr ( least squares regression ) , glr ( generalized linear regression ) , mvlr ( multivariate linear regression ) , pslr ( partial least squares regression ) , grnn ( generalized regression neural network ) , mlp ( multilayer perceptron ) and svr ( support vector regression ) are used', ['machine learning algorithms', 'regression trees'])\n", "('method used for task', 'as the XXXXX , cart ( classification and regression trees ) , lsr ( least squares regression ) , glr ( generalized XXXXX ) , mvlr ( multivariate XXXXX ) , pslr ( partial least squares regression ) , grnn ( generalized regression neural network ) , mlp ( multilayer perceptron ) and svr ( support vector regression ) are used', ['machine learning algorithms', 'linear regression'])\n", "('method used for task', 'as the XXXXX , cart ( classification and regression trees ) , lsr ( least squares regression ) , glr ( generalized XXXXX ) , mvlr ( multivariate XXXXX ) , pslr ( partial least squares regression ) , grnn ( generalized regression neural network ) , mlp ( multilayer perceptron ) and svr ( support vector regression ) are used', ['machine learning algorithms', 'linear regression'])\n", "('method used for task', 'as the XXXXX , cart ( classification and regression trees ) , lsr ( least squares regression ) , glr ( generalized linear regression ) , mvlr ( multivariate linear regression ) , pslr ( partial least squares regression ) , grnn ( generalized regression neural network ) , mlp ( XXXXX ) and svr ( support vector regression ) are used', ['machine learning algorithms', 'multilayer perceptron'])\n", "('method used for task', 'as the XXXXX , cart ( classification and regression trees ) , lsr ( least squares regression ) , glr ( generalized linear regression ) , mvlr ( multivariate linear regression ) , pslr ( partial least squares regression ) , grnn ( generalized regression neural network ) , mlp ( multilayer perceptron ) and svr ( XXXXX ) are used', ['machine learning algorithms', 'support vector regression'])\n", "('method used for task', 'as the XXXXX , cart ( classification and regression trees ) , lsr ( least squares regression ) , glr ( generalized linear regression ) , mvlr ( multivariate linear regression ) , pslr ( partial least squares regression ) , grnn ( generalized XXXXX ) , mlp ( multilayer perceptron ) and svr ( support vector regression ) are used', ['machine learning algorithms', 'regression neural network'])\n", "('NONE', 'as the machine learning algorithms , cart ( classification and XXXXX ) , lsr ( least squares regression ) , glr ( generalized XXXXX ) , mvlr ( multivariate XXXXX ) , pslr ( partial least squares regression ) , grnn ( generalized regression neural network ) , mlp ( multilayer perceptron ) and svr ( support vector regression ) are used', ['regression trees', 'linear regression'])\n", "('NONE', 'as the machine learning algorithms , cart ( classification and XXXXX ) , lsr ( least squares regression ) , glr ( generalized XXXXX ) , mvlr ( multivariate XXXXX ) , pslr ( partial least squares regression ) , grnn ( generalized regression neural network ) , mlp ( multilayer perceptron ) and svr ( support vector regression ) are used', ['regression trees', 'linear regression'])\n", "('NONE', 'as the machine learning algorithms , cart ( classification and XXXXX ) , lsr ( least squares regression ) , glr ( generalized linear regression ) , mvlr ( multivariate linear regression ) , pslr ( partial least squares regression ) , grnn ( generalized regression neural network ) , mlp ( XXXXX ) and svr ( support vector regression ) are used', ['regression trees', 'multilayer perceptron'])\n", "('method used for task', 'as the machine learning algorithms , cart ( classification and XXXXX ) , lsr ( least squares regression ) , glr ( generalized linear regression ) , mvlr ( multivariate linear regression ) , pslr ( partial least squares regression ) , grnn ( generalized regression neural network ) , mlp ( multilayer perceptron ) and svr ( XXXXX ) are used', ['regression trees', 'support vector regression'])\n", "('NONE', 'as the machine learning algorithms , cart ( classification and XXXXX ) , lsr ( least squares regression ) , glr ( generalized linear regression ) , mvlr ( multivariate linear regression ) , pslr ( partial least squares regression ) , grnn ( generalized XXXXX ) , mlp ( multilayer perceptron ) and svr ( support vector regression ) are used', ['regression trees', 'regression neural network'])\n", "('NONE', 'as the machine learning algorithms , cart ( classification and regression trees ) , lsr ( least squares regression ) , glr ( generalized XXXXX ) , mvlr ( multivariate XXXXX ) , pslr ( partial least squares regression ) , grnn ( generalized regression neural network ) , mlp ( XXXXX ) and svr ( support vector regression ) are used', ['linear regression', 'multilayer perceptron'])\n", "('NONE', 'as the machine learning algorithms , cart ( classification and regression trees ) , lsr ( least squares regression ) , glr ( generalized XXXXX ) , mvlr ( multivariate XXXXX ) , pslr ( partial least squares regression ) , grnn ( generalized regression neural network ) , mlp ( multilayer perceptron ) and svr ( XXXXX ) are used', ['linear regression', 'support vector regression'])\n", "('NONE', 'as the machine learning algorithms , cart ( classification and regression trees ) , lsr ( least squares regression ) , glr ( generalized XXXXX ) , mvlr ( multivariate XXXXX ) , pslr ( partial least squares regression ) , grnn ( generalized XXXXX ) , mlp ( multilayer perceptron ) and svr ( support vector regression ) are used', ['linear regression', 'regression neural network'])\n", "('NONE', 'as the machine learning algorithms , cart ( classification and regression trees ) , lsr ( least squares regression ) , glr ( generalized XXXXX ) , mvlr ( multivariate XXXXX ) , pslr ( partial least squares regression ) , grnn ( generalized regression neural network ) , mlp ( XXXXX ) and svr ( support vector regression ) are used', ['linear regression', 'multilayer perceptron'])\n", "('NONE', 'as the machine learning algorithms , cart ( classification and regression trees ) , lsr ( least squares regression ) , glr ( generalized XXXXX ) , mvlr ( multivariate XXXXX ) , pslr ( partial least squares regression ) , grnn ( generalized regression neural network ) , mlp ( multilayer perceptron ) and svr ( XXXXX ) are used', ['linear regression', 'support vector regression'])\n", "('NONE', 'as the machine learning algorithms , cart ( classification and regression trees ) , lsr ( least squares regression ) , glr ( generalized XXXXX ) , mvlr ( multivariate XXXXX ) , pslr ( partial least squares regression ) , grnn ( generalized XXXXX ) , mlp ( multilayer perceptron ) and svr ( support vector regression ) are used', ['linear regression', 'regression neural network'])\n", "('method used for task', 'as the machine learning algorithms , cart ( classification and regression trees ) , lsr ( least squares regression ) , glr ( generalized linear regression ) , mvlr ( multivariate linear regression ) , pslr ( partial least squares regression ) , grnn ( generalized regression neural network ) , mlp ( XXXXX ) and svr ( XXXXX ) are used', ['multilayer perceptron', 'support vector regression'])\n", "('NONE', 'as the machine learning algorithms , cart ( classification and regression trees ) , lsr ( least squares regression ) , glr ( generalized linear regression ) , mvlr ( multivariate linear regression ) , pslr ( partial least squares regression ) , grnn ( generalized XXXXX ) , mlp ( XXXXX ) and svr ( support vector regression ) are used', ['multilayer perceptron', 'regression neural network'])\n", "('method used for task', 'as the machine learning algorithms , cart ( classification and regression trees ) , lsr ( least squares regression ) , glr ( generalized linear regression ) , mvlr ( multivariate linear regression ) , pslr ( partial least squares regression ) , grnn ( generalized XXXXX ) , mlp ( multilayer perceptron ) and svr ( XXXXX ) are used', ['support vector regression', 'regression neural network'])\n", "('method used for task', 'a surrogate based on XXXXX is adapted for mixed-type variables , XXXXX and constraints and integrated into nsga-ii', ['radial basis function networks', 'multiple objectives'])\n", "('method used for task', 'variants of nsga-ii including these changes are applied to a typical building optimisation problem , with improvements in XXXXX and XXXXX', ['convergence speed', 'solution quality'])\n", "('method used for task', 'proposing XXXXX to raise the capability for XXXXX and raise the system accuracy', ['fuzzy inference system', 'fault tolerance'])\n", "('NONE', 'the on-line adaptation mechanism of neurocontroller weights relies on levenberg–marquardt method with XXXXX for adaptive changes of a XXXXX', ['fuzzy model', 'learning rate'])\n", "('NONE', 'the developed speed neurocontroller uses only easy measurable XXXXX ( no sensors or estimators of other mechanical XXXXX are applied )', ['motor speed', 'state variables'])\n", "('method used for task', 'XXXXX optimized online anfis based speed controller presented for XXXXX', ['bat algorithm', 'brushless dc motor'])\n", "('method used for task', 'XXXXX optimized online anfis based XXXXX presented for brushless dc motor', ['bat algorithm', 'speed controller'])\n", "('method used for task', 'bat algorithm optimized online anfis based XXXXX presented for XXXXX', ['brushless dc motor', 'speed controller'])\n", "('NONE', 'a mixed XXXXX ( mip ) , a XXXXX and four meta-heuristics are developed', ['local search', 'integer model'])\n", "('method used for task', 'developing algorithms to extract training patterns from XXXXX for constructing XXXXX', ['historical data', 'forecasting models'])\n", "('method used for task', 'the ea-bnb achieved improvements regarding both XXXXX and XXXXX', ['cpu time', 'solution quality'])\n", "('NONE', 'we propose an asynchronous XXXXX of XXXXX ( de )', ['parallel implementation', 'differential evolution'])\n", "('NONE', 'tested with benchmarks problems , including the calibration of non-linear XXXXX of XXXXX', ['dynamic models', 'biological systems'])\n", "('method used for task', 'the XXXXX influence the population space to guide the feasible XXXXX', ['search space', 'knowledge sources'])\n", "('method used for task', 'contribution : hybridizing XXXXX and XXXXX', ['linear programming', 'genetic algorithm'])\n", "('method used for task', 'XXXXX is preserved by proposed XXXXX algorithm', ['image quality', 'pixel selection'])\n", "('NONE', 'XXXXX which include XXXXX , feature selection and classification for determining the disc abnormalities of the lumbar region have been developed', ['hybrid models', 'feature extraction'])\n", "('NONE', 'XXXXX which include feature extraction , XXXXX and classification for determining the disc abnormalities of the lumbar region have been developed', ['hybrid models', 'feature selection'])\n", "('method used for task', 'XXXXX which include feature extraction , feature selection and classification for determining the XXXXX of the lumbar region have been developed', ['hybrid models', 'disc abnormalities'])\n", "('NONE', 'hybrid models which include XXXXX , XXXXX and classification for determining the disc abnormalities of the lumbar region have been developed', ['feature extraction', 'feature selection'])\n", "('method used for task', 'hybrid models which include XXXXX , feature selection and classification for determining the XXXXX of the lumbar region have been developed', ['feature extraction', 'disc abnormalities'])\n", "('method used for task', 'hybrid models which include feature extraction , XXXXX and classification for determining the XXXXX of the lumbar region have been developed', ['feature selection', 'disc abnormalities'])\n", "('NONE', 'the results obtained suggest that the proposed XXXXX can be safely used in the detection of XXXXX', ['hybrid models', 'disc abnormalities'])\n", "('NONE', 'the model utilizes XXXXX patterns and socio-economic data in the XXXXX data', ['activity sequence', 'time use'])\n", "('NONE', 'the XXXXX of different XXXXX are examined', ['classification methods', 'classification performances'])\n", "('method used for task', 'we modify the XXXXX based on prefetching to fully exploit the potential map tasks with XXXXX in section 4.3.1. this method has the advantages of reducing network transmission . furthermore , we consider part of nodes , whose remaining time is less then threshold tunder to avoid invalid data prefetching', ['scheduling algorithm', 'data locality'])\n", "('NONE', 'age and sms-emoa typically achieve the XXXXX of the true XXXXX', ['best approximation', 'pareto front'])\n", "('method used for task', 'XXXXX : adjust parameters and XXXXX to the latest information', ['incremental learning', 'model structure'])\n", "('NONE', 'the proposed method is a XXXXX that combines XXXXX and k-means', ['hybrid method', 'the neural networks'])\n", "('method used for task', 'XXXXX with random small steps ( exploitation ) is performed only if XXXXX fails', ['local search', 'global search'])\n", "('method used for task', 'XXXXX to XXXXX ( odes ) in engineering', ['approximate solutions', 'ordinary differential equations'])\n", "('method used for task', 'we create a XXXXX to solve the XXXXX', ['modified differential evolution', 'optimization problem'])\n", "('method used for task', 'a new XXXXX is developed to reduce XXXXX for vehicle routing', ['mathematical model', 'co2 emissions'])\n", "('method used for task', 'a new XXXXX is developed to reduce co2 emissions for XXXXX', ['mathematical model', 'vehicle routing'])\n", "('method used for task', 'a new mathematical model is developed to reduce XXXXX for XXXXX', ['co2 emissions', 'vehicle routing'])\n", "('NONE', 'XXXXX can be reduced by taking traffic congestion into account in XXXXX', ['co2 emissions', 'vehicle routes'])\n", "('NONE', 'a soft computing method is firstly used in the literature to determine the XXXXX of tissue equivalent liquid exposed to XXXXX', ['temperature distribution', 'electromagnetic field'])\n", "('method used for task', 'a XXXXX is firstly used in the literature to determine the XXXXX of tissue equivalent liquid exposed to electromagnetic field', ['temperature distribution', 'soft computing method'])\n", "('method used for task', 'a XXXXX is firstly used in the literature to determine the temperature distribution of tissue equivalent liquid exposed to XXXXX', ['electromagnetic field', 'soft computing method'])\n", "('method used for task', 'a switching adaptive control scheme using a hopfield-based dynamic XXXXX ( sachnn ) for XXXXX with external disturbances is proposed', ['neural network', 'nonlinear systems'])\n", "('method used for task', 'we propose a model which incorporates kernel metric and XXXXX for XXXXX', ['fuzzy logic', 'image segmentation'])\n", "('NONE', 'nox and soot emissions from a XXXXX are modeled using XXXXX', ['diesel engine', 'artificial neural network model'])\n", "('NONE', 'the data used for training and testing the XXXXX are obtained through testing a XXXXX', ['diesel engine', 'ann model'])\n", "('NONE', 'the mass fuel rate , intake XXXXX , etc . are optimized using XXXXX', ['air temperature', 'ant colony optimization algorithm'])\n", "('NONE', 'a cheap and portable approach to detect XXXXX in XXXXX is proposed', ['fall detection', 'real time'])\n", "('method used for task', 'XXXXX are gathered by a wearable sensor and sent to a XXXXX', ['mobile device', 'acceleration data'])\n", "('NONE', 'provide an updated and XXXXX of distributed XXXXX', ['systematic review', 'evolutionary algorithms'])\n", "('NONE', 'location-routing XXXXX including XXXXX and direct shipment is modeled', ['scheduling problem', 'cross docking'])\n", "('method used for task', 'by the use of the bi-clustering method , XXXXX are solved by XXXXX', ['large-scale problems', 'exact methods'])\n", "('method used for task', 'XXXXX for modular robots equipped with XXXXX in complex environments', ['motion planning', 'motion primitives'])\n", "('method used for task', 'novel method for adaptation of the XXXXX based on XXXXX is proposed', ['particle swarm optimization', 'motion primitives'])\n", "('NONE', 'to reduce the impact of XXXXX , we incorporate XXXXX into antibody–antigen affinity', ['speckle noise', 'local information'])\n", "('method used for task', 'a new XXXXX is proposed for the jso that incorporates the XXXXX in the local search procedure', ['memetic algorithm', 'neighborhood structures'])\n", "('method used for task', 'a new XXXXX is proposed for the jso that incorporates the neighborhood structures in the XXXXX', ['memetic algorithm', 'local search procedure'])\n", "('NONE', 'a new memetic algorithm is proposed for the jso that incorporates the XXXXX in the XXXXX', ['neighborhood structures', 'local search procedure'])\n", "('NONE', 'an XXXXX was conducted showing that the XXXXX compares favorably with the state-of-the-art', ['experimental study', 'memetic algorithm'])\n", "('NONE', 'incorporation of XXXXX into XXXXX could be beneficial to text categorization task', ['topic identification', 'som learning'])\n", "('method used for task', 'two layered cascade-based XXXXX are proposed for XXXXX', ['evolutionary algorithms', 'feature selection'])\n", "('method used for task', 'both a XXXXX and an adaptive XXXXX are employed for expression recognition', ['neural network', 'ensemble classifier'])\n", "('method used for task', 'both a XXXXX and an adaptive ensemble classifier are employed for XXXXX', ['neural network', 'expression recognition'])\n", "('method used for task', 'both a neural network and an adaptive XXXXX are employed for XXXXX', ['ensemble classifier', 'expression recognition'])\n", "('method used for task', 'the orthogonal XXXXX is integrated into XXXXX', ['differential evolution', 'initialization method'])\n", "('method used for task', 'modified XXXXX is used to control XXXXX with simulated annealing technique', ['mutation operator', 'convergence rate'])\n", "('NONE', 'our XXXXX can only use XXXXX to mitigate the imbalance of the loads', ['distributed algorithm', 'local information'])\n", "('method used for task', 'a new color image segmentation method based on improved bee XXXXX for XXXXX ( ibmo ) is presented', ['multi-objective optimization', 'colony algorithm'])\n", "('NONE', 'the obtained results are compared the ones obtained from XXXXX which is one of the most popular methods used in XXXXX , nondominated sorting genetic algorithm , and nondominated sorted particle swarm optimization', ['fuzzy c-means', 'image segmentation'])\n", "('NONE', 'the obtained results are compared the ones obtained from XXXXX which is one of the most popular methods used in image segmentation , nondominated sorting XXXXX , and nondominated sorted particle swarm optimization', ['fuzzy c-means', 'genetic algorithm'])\n", "('method used for task', 'the obtained results are compared the ones obtained from XXXXX which is one of the most popular methods used in image segmentation , nondominated sorting genetic algorithm , and nondominated sorted XXXXX', ['fuzzy c-means', 'particle swarm optimization'])\n", "('NONE', 'the obtained results are compared the ones obtained from fuzzy c-means which is one of the most popular methods used in XXXXX , nondominated sorting XXXXX , and nondominated sorted particle swarm optimization', ['image segmentation', 'genetic algorithm'])\n", "('method used for task', 'the obtained results are compared the ones obtained from fuzzy c-means which is one of the most popular methods used in XXXXX , nondominated sorting genetic algorithm , and nondominated sorted XXXXX', ['image segmentation', 'particle swarm optimization'])\n", "('method used for task', 'the obtained results are compared the ones obtained from fuzzy c-means which is one of the most popular methods used in image segmentation , nondominated sorting XXXXX , and nondominated sorted XXXXX', ['genetic algorithm', 'particle swarm optimization'])\n", "('NONE', 'better performance than existing XXXXX and other approaches such as XXXXX and fcm clustering', ['k-means clustering', 'cosine similarity'])\n", "('NONE', 'pca proved effective in creating a XXXXX of XXXXX', ['digital signature', 'network traffic'])\n", "('method used for task', 'results pertaining to XXXXX and XXXXX are encouraging', ['false alarm', 'accuracy rate'])\n", "('NONE', 'flexibility to fix the XXXXX in h∞ loop shaping XXXXX', ['controller design', 'controller structure'])\n", "('method used for task', 'proposing a XXXXX based on gbmo for XXXXX', ['hybrid algorithm', 'image classification'])\n", "('method used for task', 'non-linearity between XXXXX ( rss ) and distance is modeled using fuzzy logic system ( fls ) to reduce the XXXXX and further optimized by hpso and bbo to minimize the error', ['received signal strength', 'computational complexity'])\n", "('NONE', 'due to high XXXXX of simulations , kriging XXXXX are built', ['computational cost', 'surrogate models'])\n", "('NONE', 'particle swarm algorithm with adaptive XXXXX optimizes XXXXX', ['constraint handling', 'surrogate model'])\n", "('NONE', 'XXXXX with adaptive XXXXX optimizes surrogate model', ['constraint handling', 'particle swarm algorithm'])\n", "('NONE', 'XXXXX with adaptive constraint handling optimizes XXXXX', ['surrogate model', 'particle swarm algorithm'])\n", "('method used for task', 'the XXXXX relies on human intuition and XXXXX', ['k-means clustering', 'industry standard'])\n", "('method used for task', 'a XXXXX ( ma-sat ) is proposed for XXXXX in networks', ['memetic algorithm', 'community detection'])\n", "('NONE', 'effective XXXXX in selecting the relevant XXXXX of cancer subtypes', ['clustering techniques', 'gene expression'])\n", "('NONE', 'it combines XXXXX with estimation of XXXXX', ['particle swarm optimization', 'distribution algorithm'])\n", "('NONE', 'a conditional spatial XXXXX ( csfcm ) XXXXX to improve the robustness of the conventional fcm algorithm is presented', ['fuzzy c-means', 'clustering algorithm'])\n", "('NONE', 'a conditional spatial XXXXX ( csfcm ) clustering algorithm to improve the robustness of the conventional XXXXX is presented', ['fuzzy c-means', 'fcm algorithm'])\n", "('NONE', 'a conditional spatial fuzzy c-means ( csfcm ) XXXXX to improve the robustness of the conventional XXXXX is presented', ['clustering algorithm', 'fcm algorithm'])\n", "('NONE', 'the method incorporates conditional affects and XXXXX into the XXXXX', ['spatial information', 'membership functions'])\n", "('NONE', 'this algorithm is based on the full XXXXX of XXXXX', ['bayesian approach', 'artificial neural networks'])\n", "('method used for task', 'XXXXX are integrated with gas and XXXXX', ['monte carlo methods', 'fuzzy membership functions'])\n", "('method used for task', 'proposed algorithm is applied to XXXXX and XXXXX in the context of bnns', ['time series', 'regression analysis'])\n", "('method used for task', 'XXXXX and XXXXX are roped as a level two classifier for good classification accuracy', ['nonlinear models', 'neural networks'])\n", "('method used for task', 'XXXXX and neural networks are roped as a level two classifier for good XXXXX', ['nonlinear models', 'classification accuracy'])\n", "('method used for task', 'nonlinear models and XXXXX are roped as a level two classifier for good XXXXX', ['neural networks', 'classification accuracy'])\n", "('method used for task', 'the actual XXXXX is carried out by XXXXX', ['time series prediction', 'neural networks'])\n", "('NONE', 'we integrate the XXXXX module into the neverstop XXXXX . the XXXXX architecture completes the integration and modeling of the traffic control systems', ['system design', 'fuzzy control'])\n", "('NONE', 'we integrate the XXXXX into the neverstop XXXXX . the fuzzy control architecture completes the integration and modeling of the traffic control systems', ['system design', 'fuzzy control module'])\n", "('NONE', 'we integrate the XXXXX module into the neverstop system design . the XXXXX architecture completes the integration and modeling of the traffic control systems', ['fuzzy control', 'fuzzy control module'])\n", "('NONE', 'we present a XXXXX illustrating how the average XXXXX is derived . the involvement amplifies the neverstop system and facilitates the fuzzy control module', ['genetic algorithm', 'waiting time'])\n", "('method used for task', 'we present a XXXXX illustrating how the average waiting time is derived . the involvement amplifies the neverstop system and facilitates the XXXXX', ['genetic algorithm', 'fuzzy control module'])\n", "('method used for task', 'we present a genetic algorithm illustrating how the average XXXXX is derived . the involvement amplifies the neverstop system and facilitates the XXXXX', ['waiting time', 'fuzzy control module'])\n", "('method used for task', 'neverstop utilizes fuzzy control method and XXXXX to adjust the XXXXX for the traffic lights , consequently the average XXXXX can be significantly reduced', ['genetic algorithm', 'waiting time'])\n", "('method used for task', 'neverstop utilizes fuzzy control method and XXXXX to adjust the XXXXX for the traffic lights , consequently the average XXXXX can be significantly reduced', ['genetic algorithm', 'waiting time'])\n", "('method used for task', 'neverstop utilizes fuzzy control method and XXXXX to adjust the waiting time for the XXXXX , consequently the average waiting time can be significantly reduced', ['genetic algorithm', 'traffic lights'])\n", "('method used for task', 'neverstop utilizes fuzzy control method and genetic algorithm to adjust the XXXXX for the XXXXX , consequently the average XXXXX can be significantly reduced', ['waiting time', 'traffic lights'])\n", "('method used for task', 'neverstop utilizes fuzzy control method and genetic algorithm to adjust the XXXXX for the XXXXX , consequently the average XXXXX can be significantly reduced', ['waiting time', 'traffic lights'])\n", "('NONE', 'the XXXXX is enhanced in the proposed system for fast retrieval of hidden data from XXXXX', ['hidden markov model', 'video files'])\n", "('method used for task', 'XXXXX and retrieval processes are performed using the conditional states and state transition dynamics between the XXXXX', ['data embedding', 'video frames'])\n", "('NONE', 'it enhances the retrieval XXXXX with minimized XXXXX', ['data rate', 'computation cost'])\n", "('method used for task', 'this paper proposes a new XXXXX mechanism for basic XXXXX', ['abc algorithm', 'solution update'])\n", "('method used for task', 'the XXXXX rule is based on XXXXX', ['normal distribution', 'solution update'])\n", "('method used for task', 'a novel XXXXX , coa-ga , is developed by integrating cuckoo XXXXX ( coa ) and ga to enhance XXXXX', ['optimization algorithm', 'classification performance'])\n", "('method used for task', 'a novel XXXXX , coa-ga , is developed by integrating cuckoo XXXXX ( coa ) and ga to enhance XXXXX', ['optimization algorithm', 'classification performance'])\n", "('method used for task', 'it is further confirmed that traditional clustering does not have any impact on XXXXX and XXXXX', ['gene selection', 'classification performance'])\n", "('NONE', 'adaptive XXXXX selects the most appropriate operators in an XXXXX', ['evolutionary algorithm', 'operator selection'])\n", "('NONE', 'hybridizes the XXXXX with XXXXX , where XXXXX is applied to control the randomness step inside the XXXXX', ['firefly algorithm', 'simulated annealing'])\n", "('NONE', 'hybridizes the XXXXX with XXXXX , where XXXXX is applied to control the randomness step inside the XXXXX', ['firefly algorithm', 'simulated annealing'])\n", "('NONE', 'hybridizes the XXXXX with XXXXX , where XXXXX is applied to control the randomness step inside the XXXXX', ['simulated annealing', 'firefly algorithm'])\n", "('NONE', 'hybridizes the XXXXX with XXXXX , where XXXXX is applied to control the randomness step inside the XXXXX', ['simulated annealing', 'firefly algorithm'])\n", "('method used for task', 'a XXXXX is embedded within the XXXXX to better explore the search space', ['lévy flight', 'firefly algorithm'])\n", "('method used for task', 'a XXXXX is embedded within the firefly algorithm to better explore the XXXXX', ['lévy flight', 'search space'])\n", "('method used for task', 'a lévy flight is embedded within the XXXXX to better explore the XXXXX', ['firefly algorithm', 'search space'])\n", "('method used for task', 'a combination of firefly , XXXXX and XXXXX is investigated to further improve the solution', ['lévy flight', 'simulated annealing'])\n", "('method used for task', 'the XXXXX is integrated in the conventional XXXXX', ['similarity measure', 'fuzzy c-means algorithm'])\n", "('method used for task', 'a compilation of XXXXX approaches for solving the XXXXX', ['soft computing', 'protein structure prediction problem'])\n", "('NONE', 'a summary of the basic concepts in the XXXXX of XXXXX', ['protein structure prediction', 'research field'])\n", "('NONE', 'in this study , an urban XXXXX dataset received from the database of uci XXXXX was used as the urban XXXXX data', ['machine learning', 'land cover'])\n", "('method used for task', 'this paper aims to improve the performance of XXXXX for solving XXXXX', ['harmony search algorithm', 'multi-objective optimization problems'])\n", "('method used for task', 'secondly , the proposed algorithm is applied to solve many classical XXXXX and it is also compared with other XXXXX', ['benchmark problems', 'multi-objective evolutionary algorithms'])\n", "('NONE', 'XXXXX experiments show that the controller proposed ( based on XXXXX ) is stable', ['fuzzy logic', 'lane changes'])\n", "('NONE', 'three XXXXX namely , iterative bi-section shooting ( ibss ) , XXXXX ( ann ) , and the novel new hybrid ann–ibss , were proposed', ['artificial neural network', 'parameter estimation methods'])\n", "('NONE', 'this paper has presented the our new XXXXX with all results in qos ( quality of XXXXX ) of wsn ( wireless sensor nodes ) and it is compared with leach , sep , genetic hcr and erp routing protocols', ['routing protocol', 'service metrics'])\n", "('NONE', 'three XXXXX are considered : XXXXX , power loss , voltage deviation and system loadability', ['objective functions', 'fuel cost'])\n", "('NONE', 'three XXXXX are considered : fuel cost , XXXXX , voltage deviation and system loadability', ['objective functions', 'power loss'])\n", "('NONE', 'three XXXXX are considered : fuel cost , power loss , XXXXX and system loadability', ['objective functions', 'voltage deviation'])\n", "('NONE', 'three objective functions are considered : XXXXX , XXXXX , voltage deviation and system loadability', ['fuel cost', 'power loss'])\n", "('NONE', 'three objective functions are considered : XXXXX , power loss , XXXXX and system loadability', ['fuel cost', 'voltage deviation'])\n", "('NONE', 'three objective functions are considered : fuel cost , XXXXX , XXXXX and system loadability', ['power loss', 'voltage deviation'])\n", "('NONE', 'modified XXXXX using XXXXX and external archive for multi-objective optimization', ['bat algorithm', 'inertia weight'])\n", "('method used for task', 'modified XXXXX using inertia weight and external archive for XXXXX', ['bat algorithm', 'multi-objective optimization'])\n", "('method used for task', 'modified bat algorithm using XXXXX and external archive for XXXXX', ['inertia weight', 'multi-objective optimization'])\n", "('method used for task', \"single-pass XXXXX for reducing operator 's annotation effort during on-line XXXXX\", ['active learning', 'visual inspection'])\n", "('method used for task', \"single-pass XXXXX for reducing operator 's XXXXX during on-line visual inspection\", ['active learning', 'annotation effort'])\n", "('NONE', \"single-pass active learning for reducing operator 's XXXXX during on-line XXXXX\", ['visual inspection', 'annotation effort'])\n", "('method used for task', 'intelligent hardware is based on the XXXXX and XXXXX ( nn )', ['wavelet decomposition', 'neural network'])\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "('NONE', 'integration of XXXXX with XXXXX for policy iteration', ['reinforcement learning', 'gaussian processes'])\n", "('method used for task', 'based on the adaptive neuro-fuzzy XXXXX ( anfis ) approach , a hazard modeling and survival prediction system is developed to assist clinicians in prognostic assessment of patients with esophageal cancer and prediction of individual XXXXX', ['inference system', 'patient survival'])\n", "('method used for task', 'this may provide valuable prognostic information in addition to ajcc staging and aid the clinicians’ XXXXX for XXXXX', ['decision-making process', 'risk stratification'])\n", "('method used for task', 'ann shows good XXXXX as compared to svr , rfr and XXXXX', ['prediction performance', 'mlr models'])\n", "('method used for task', 'this type of XXXXX can help to represent XXXXX having certain order', ['soft sets', 'linguistic terms'])\n", "('NONE', 'in certain XXXXX these XXXXX and operations on them can be very helpful', ['soft sets', 'decision making problems'])\n", "('NONE', 'incorporation of XXXXX in single- and three-area XXXXX', ['pid controller', 'power systems'])\n", "('NONE', 'investigation of idd XXXXX in five-area XXXXX', ['power system', 'controller performance'])\n", "('method used for task', 'a novel profit-based XXXXX for XXXXX with svm is presented', ['churn prediction', 'feature selection method'])\n", "('method used for task', 'determining critical XXXXX rules and forming a flexible XXXXX for XXXXX', ['rule base', 'phishing detection'])\n", "('method used for task', 'determining critical XXXXX rules and forming a flexible XXXXX for XXXXX', ['rule base', 'phishing detection'])\n", "('NONE', 'this study introduces a new methodology to practitioners and academics in which XXXXX used , in the XXXXX using the efqm excellence model for dealing with imprecision and applying ahp & or in the priority setting of improvement projects', ['fuzzy logic', 'performance evaluation'])\n", "('NONE', 'this study introduces a new methodology to practitioners and academics in which XXXXX used , in the performance evaluation using the efqm excellence model for dealing with imprecision and applying ahp & or in the priority setting of XXXXX', ['fuzzy logic', 'improvement projects'])\n", "('NONE', 'this study introduces a new methodology to practitioners and academics in which fuzzy logic used , in the XXXXX using the efqm excellence model for dealing with imprecision and applying ahp & or in the priority setting of XXXXX', ['performance evaluation', 'improvement projects'])\n", "('NONE', 'first one is the use of XXXXX for efqm excellence model not only make easier the procedure of XXXXX but also the total score for the organization performance obtained will be closer to the real score', ['fuzzy logic', 'performance assessment'])\n", "('method used for task', 'combining asynchronously XXXXX and XXXXX is the best program', ['inertia weight', 'constriction factor'])\n", "('method used for task', 'XXXXX for automatic XXXXX repudiate the need of a human expert', ['decision trees', 'rule learning'])\n", "('method used for task', 'automatic fuzzy partitioning of the XXXXX to over XXXXX', ['feature space', 'linguistic terms'])\n", "('NONE', 'high true positive rate and lower false positive rate for the proposed fuzzy rule base classification . high XXXXX achieved by the proposed system over XXXXX', ['classification accuracy', 'learning algorithms'])\n", "('method used for task', 'an ecg heart beat XXXXX is proposed based on modified XXXXX', ['abc algorithm', 'classification method'])\n", "('method used for task', 'total 38 XXXXX is calculated , then most distinctive XXXXX is used', ['feature set', 'feature subset'])\n", "('method used for task', 'XXXXX is achieved as 99.30 % on examined XXXXX from mitbih db', ['classification accuracy', 'ecg data'])\n", "('NONE', 'XXXXX within the interval type-2 XXXXX', ['multiple criteria decision analysis', 'fuzzy environment'])\n", "('method used for task', 'XXXXX are designed to test the XXXXX', ['computational experiments', 'solution approach'])\n", "('method used for task', 'XXXXX hp-hgs for solving XXXXX', ['hybrid strategy', 'inverse problems'])\n", "('NONE', 'global step solved by hierarchical XXXXX ( hgs ) coupled with self-adaptive hp-finite XXXXX ( hp-fem )', ['genetic search', 'element method'])\n", "('method used for task', 'rule pools for XXXXX are generated by a rule-based XXXXX', ['stock trading', 'evolutionary algorithm'])\n", "('method used for task', 'XXXXX for XXXXX are generated by a rule-based evolutionary algorithm', ['stock trading', 'rule pools'])\n", "('method used for task', 'XXXXX for stock trading are generated by a rule-based XXXXX', ['evolutionary algorithm', 'rule pools'])\n", "('NONE', 'XXXXX using mlp selects appropriate XXXXX for trading decision', ['ensemble learning', 'rule pools'])\n", "('NONE', 'properties of the fuzzy linguistic XXXXX on discrete XXXXX are analysed', ['model based', 'fuzzy numbers'])\n", "('NONE', 'some advantages of the XXXXX on discrete XXXXX are pointed out', ['model based', 'fuzzy numbers'])\n", "('NONE', 'aiming at the chaotic behavior of pmsg under certain parameters , a method based on adhdp is proposed to track the maximum wind power point , which can make the system out of chaos and track the maximum power point stably , and the XXXXX of the complex XXXXX can be solved effectively', ['optimal control problems', 'nonlinear system'])\n", "('method used for task', 'neutrosophic XXXXX is proposed based on XXXXX', ['cost function', 'neutrosophic set'])\n", "('NONE', 'solving the XXXXX in structure designing and XXXXX', ['parameter optimization', 'machining process'])\n", "('NONE', 'the proposed algorithm has better XXXXX than several popular XXXXX', ['computational efficiency', 'soft computing algorithms'])\n", "('method used for task', 'considering new features namely , XXXXX , opening and reopening modes , XXXXX and transportation modes', ['production facilities', 'greenhouse gas'])\n", "('method used for task', 'considering new features namely , XXXXX , opening and reopening modes , greenhouse gas and XXXXX', ['production facilities', 'transportation modes'])\n", "('method used for task', 'considering new features namely , production facilities , opening and reopening modes , XXXXX and XXXXX', ['greenhouse gas', 'transportation modes'])\n", "('NONE', 'to improve the exploration and XXXXX of bat algorithm some modifications based on XXXXX on bat algorithm are studied', ['chaotic map', 'exploitation capability'])\n", "('NONE', 'a new XXXXX inspired by XXXXX was proposed', ['tumor growth', 'heuristic optimization algorithm'])\n", "('method used for task', 'an empirical XXXXX is presented for solving different XXXXX', ['comparison study', 'test functions'])\n", "('method used for task', 'formally describes a form of modeling that unifies XXXXX and XXXXX', ['agent-based modeling', 'complex networks'])\n", "('NONE', 'we propose a method to transform the lineage expression into directed acyclic graphs ( dags ) equivalently starting from the lineage expressed as boolean formulas for spj queries over XXXXX . specifically , we discuss the corresponding XXXXX and properties to guarantee that the graphical model can support effective probabilistic inferences in lineage processing theoretically', ['uncertain data', 'probabilistic semantics'])\n", "('method used for task', 'we propose a method to transform the lineage expression into directed acyclic graphs ( dags ) equivalently starting from the lineage expressed as boolean formulas for spj queries over XXXXX . specifically , we discuss the corresponding probabilistic semantics and properties to guarantee that the XXXXX can support effective probabilistic inferences in lineage processing theoretically', ['uncertain data', 'graphical model'])\n", "('method used for task', 'we propose a method to transform the lineage expression into directed acyclic graphs ( dags ) equivalently starting from the lineage expressed as boolean formulas for spj queries over uncertain data . specifically , we discuss the corresponding XXXXX and properties to guarantee that the XXXXX can support effective probabilistic inferences in lineage processing theoretically', ['probabilistic semantics', 'graphical model'])\n", "('method used for task', 'we develop an XXXXX to solve interrelated XXXXX', ['evolutionary approach', 'optimisation problems'])\n", "('method used for task', 'an exact analytical XXXXX is introduced for interval type-2 XXXXX with closed form inference methods', ['inversion procedure', 'tsk fuzzy logic systems'])\n", "('NONE', 'the proposed procedure enables the use of more freely designed interval type-2 XXXXX in exact inverse model based XXXXX', ['engineering applications', 'tsk fuzzy logic systems'])\n", "('NONE', 'a new bionic algorithm is proposed based on the XXXXX of XXXXX branching , regrowing and tropisms', ['incentive mechanism', 'plant root'])\n", "('method used for task', 'dual XXXXX transform ( dtcwt ) is used as a new feature extractor from forward and reverse XXXXX', ['doppler ultrasound signals', 'tree complex wavelet'])\n", "('method used for task', 'an approach combining XXXXX and XXXXX is proposed', ['simulated annealing', 'mathematical programming'])\n", "('NONE', 'classifier diversity is achieved by diversifying the XXXXX and varying the initialization of the weights of the XXXXX', ['neural network', 'input features'])\n", "('method used for task', 'we examine changes in output of the proposed fusion model with increasing XXXXX , for all the XXXXX', ['test cases', 'base classifiers'])\n", "('NONE', 'lipschitz XXXXX and copula characteristic of t-norms and implications are discussed and the robustness of rule-based XXXXX is investigated', ['fuzzy reasoning', 'aggregation property'])\n", "('NONE', 'lipschitz aggregation property and XXXXX of t-norms and implications are discussed and the robustness of rule-based XXXXX is investigated', ['fuzzy reasoning', 'copula characteristic'])\n", "('method used for task', 'lipschitz XXXXX and XXXXX of t-norms and implications are discussed and the robustness of rule-based fuzzy reasoning is investigated', ['aggregation property', 'copula characteristic'])\n", "('NONE', 'according to lipschitz XXXXX and copula characteristic of t-norms and implications , suitable t-norm and implication can be chosen to satisfy the need of robustness of XXXXX', ['fuzzy reasoning', 'aggregation property'])\n", "('NONE', 'according to lipschitz aggregation property and XXXXX of t-norms and implications , suitable t-norm and implication can be chosen to satisfy the need of robustness of XXXXX', ['fuzzy reasoning', 'copula characteristic'])\n", "('method used for task', 'according to lipschitz XXXXX and XXXXX of t-norms and implications , suitable t-norm and implication can be chosen to satisfy the need of robustness of fuzzy reasoning', ['aggregation property', 'copula characteristic'])\n", "('NONE', 'approach for the optimization of modular neural networks with the XXXXX using XXXXX for pattern recognition', ['gravitational search algorithm', 'fuzzy logic'])\n", "('NONE', 'approach for the optimization of modular neural networks with the XXXXX using fuzzy logic for XXXXX', ['gravitational search algorithm', 'pattern recognition'])\n", "('method used for task', 'approach for the optimization of modular neural networks with the gravitational search algorithm using XXXXX for XXXXX', ['fuzzy logic', 'pattern recognition'])\n", "('NONE', 'bespectacled XXXXX will reduce the XXXXX', ['eye image', 'iris segmentation accuracy'])\n", "('method used for task', 'XXXXX is performed as the XXXXX for frame detection', ['fuzzy logic', 'decision making'])\n", "('method used for task', 'eigenstructure based XXXXX for XXXXX', ['fault detection', 'nonlinear systems'])\n", "('method used for task', 'compared four XXXXX for learning XXXXX', ['fuzzy cognitive maps', 'stochastic optimization methods'])\n", "('method used for task', 'a novel XXXXX based neighborhood construction and a fast XXXXX', ['tree search', 'evaluation method'])\n", "('NONE', 'a novel XXXXX based XXXXX and a fast evaluation method', ['tree search', 'neighborhood construction'])\n", "('method used for task', 'a novel tree search based XXXXX and a fast XXXXX', ['evaluation method', 'neighborhood construction'])\n", "('NONE', 'the generating units’ contingencies and XXXXX are explicitly considered in the XXXXX of htss', ['price uncertainty', 'stochastic programming'])\n", "('method used for task', '18 XXXXX and two real-world problems are used in XXXXX', ['benchmark functions', 'experimental study'])\n", "('method used for task', '18 XXXXX and two XXXXX are used in experimental study', ['benchmark functions', 'real-world problems'])\n", "('NONE', '18 benchmark functions and two XXXXX are used in XXXXX', ['experimental study', 'real-world problems'])\n", "('NONE', 'presenting XXXXX in the sense of the proposed XXXXX', ['lyapunov function', 'adaptation laws'])\n", "('NONE', 'the XXXXX and the robustness of wsa were proven through a set of XXXXX', ['computational study', 'convergence capability'])\n", "('method used for task', 'an optimal XXXXX as used in dpc is prohibited due to the extremely large size of the XXXXX in this optimization problem', ['exhaustive search', 'search space'])\n", "('NONE', 'an optimal XXXXX as used in dpc is prohibited due to the extremely large size of the search space in this XXXXX', ['exhaustive search', 'optimization problem'])\n", "('NONE', 'an optimal exhaustive search as used in dpc is prohibited due to the extremely large size of the XXXXX in this XXXXX', ['search space', 'optimization problem'])\n", "('NONE', 'XXXXX ( XXXXX ) can be used as an alternative for this optimization problem to reduce the complexity of the search ( scheduling problem ) . the performance of the XXXXX with elitism and adaptive mutation is demonstrated to be near optimal as obtained with an exhaustive search . it has been demonstrated in this paper that ga achieves about 98–99 % of system sum rate as obtained with dpc with significant reduction in time and computational complexity', ['evolutionary algorithm', 'genetic algorithm'])\n", "('NONE', 'XXXXX ( genetic algorithm ) can be used as an alternative for this XXXXX to reduce the complexity of the search ( scheduling problem ) . the performance of the genetic algorithm with elitism and adaptive mutation is demonstrated to be near optimal as obtained with an exhaustive search . it has been demonstrated in this paper that ga achieves about 98–99 % of system sum rate as obtained with dpc with significant reduction in time and computational complexity', ['evolutionary algorithm', 'optimization problem'])\n", "('method used for task', 'XXXXX ( genetic algorithm ) can be used as an alternative for this optimization problem to reduce the complexity of the search ( XXXXX ) . the performance of the genetic algorithm with elitism and adaptive mutation is demonstrated to be near optimal as obtained with an exhaustive search . it has been demonstrated in this paper that ga achieves about 98–99 % of system sum rate as obtained with dpc with significant reduction in time and computational complexity', ['evolutionary algorithm', 'scheduling problem'])\n", "('NONE', 'XXXXX ( XXXXX ) can be used as an alternative for this optimization problem to reduce the complexity of the search ( scheduling problem ) . the performance of the XXXXX with elitism and adaptive mutation is demonstrated to be near optimal as obtained with an exhaustive search . it has been demonstrated in this paper that ga achieves about 98–99 % of system sum rate as obtained with dpc with significant reduction in time and computational complexity', ['evolutionary algorithm', 'genetic algorithm'])\n", "('method used for task', 'XXXXX ( genetic algorithm ) can be used as an alternative for this optimization problem to reduce the complexity of the search ( scheduling problem ) . the performance of the genetic algorithm with elitism and adaptive mutation is demonstrated to be near optimal as obtained with an XXXXX . it has been demonstrated in this paper that ga achieves about 98–99 % of system sum rate as obtained with dpc with significant reduction in time and computational complexity', ['evolutionary algorithm', 'exhaustive search'])\n", "('NONE', 'XXXXX ( genetic algorithm ) can be used as an alternative for this optimization problem to reduce the complexity of the search ( scheduling problem ) . the performance of the genetic algorithm with elitism and adaptive mutation is demonstrated to be near optimal as obtained with an exhaustive search . it has been demonstrated in this paper that ga achieves about 98–99 % of system sum rate as obtained with dpc with significant reduction in time and XXXXX', ['evolutionary algorithm', 'computational complexity'])\n", "('NONE', 'evolutionary algorithm ( XXXXX ) can be used as an alternative for this XXXXX to reduce the complexity of the search ( scheduling problem ) . the performance of the XXXXX with elitism and adaptive mutation is demonstrated to be near optimal as obtained with an exhaustive search . it has been demonstrated in this paper that ga achieves about 98–99 % of system sum rate as obtained with dpc with significant reduction in time and computational complexity', ['genetic algorithm', 'optimization problem'])\n", "('method used for task', 'evolutionary algorithm ( XXXXX ) can be used as an alternative for this optimization problem to reduce the complexity of the search ( XXXXX ) . the performance of the XXXXX with elitism and adaptive mutation is demonstrated to be near optimal as obtained with an exhaustive search . it has been demonstrated in this paper that ga achieves about 98–99 % of system sum rate as obtained with dpc with significant reduction in time and computational complexity', ['genetic algorithm', 'scheduling problem'])\n", "('NONE', 'evolutionary algorithm ( XXXXX ) can be used as an alternative for this optimization problem to reduce the complexity of the search ( scheduling problem ) . the performance of the XXXXX with elitism and adaptive mutation is demonstrated to be near optimal as obtained with an XXXXX . it has been demonstrated in this paper that ga achieves about 98–99 % of system sum rate as obtained with dpc with significant reduction in time and computational complexity', ['genetic algorithm', 'exhaustive search'])\n", "('NONE', 'evolutionary algorithm ( XXXXX ) can be used as an alternative for this optimization problem to reduce the complexity of the search ( scheduling problem ) . the performance of the XXXXX with elitism and adaptive mutation is demonstrated to be near optimal as obtained with an exhaustive search . it has been demonstrated in this paper that ga achieves about 98–99 % of system sum rate as obtained with dpc with significant reduction in time and XXXXX', ['genetic algorithm', 'computational complexity'])\n", "('NONE', 'evolutionary algorithm ( genetic algorithm ) can be used as an alternative for this XXXXX to reduce the complexity of the search ( XXXXX ) . the performance of the genetic algorithm with elitism and adaptive mutation is demonstrated to be near optimal as obtained with an exhaustive search . it has been demonstrated in this paper that ga achieves about 98–99 % of system sum rate as obtained with dpc with significant reduction in time and computational complexity', ['optimization problem', 'scheduling problem'])\n", "('NONE', 'evolutionary algorithm ( XXXXX ) can be used as an alternative for this XXXXX to reduce the complexity of the search ( scheduling problem ) . the performance of the XXXXX with elitism and adaptive mutation is demonstrated to be near optimal as obtained with an exhaustive search . it has been demonstrated in this paper that ga achieves about 98–99 % of system sum rate as obtained with dpc with significant reduction in time and computational complexity', ['optimization problem', 'genetic algorithm'])\n", "('method used for task', 'evolutionary algorithm ( genetic algorithm ) can be used as an alternative for this XXXXX to reduce the complexity of the search ( scheduling problem ) . the performance of the genetic algorithm with elitism and adaptive mutation is demonstrated to be near optimal as obtained with an XXXXX . it has been demonstrated in this paper that ga achieves about 98–99 % of system sum rate as obtained with dpc with significant reduction in time and computational complexity', ['optimization problem', 'exhaustive search'])\n", "('NONE', 'evolutionary algorithm ( genetic algorithm ) can be used as an alternative for this XXXXX to reduce the complexity of the search ( scheduling problem ) . the performance of the genetic algorithm with elitism and adaptive mutation is demonstrated to be near optimal as obtained with an exhaustive search . it has been demonstrated in this paper that ga achieves about 98–99 % of system sum rate as obtained with dpc with significant reduction in time and XXXXX', ['optimization problem', 'computational complexity'])\n", "('method used for task', 'evolutionary algorithm ( XXXXX ) can be used as an alternative for this optimization problem to reduce the complexity of the search ( XXXXX ) . the performance of the XXXXX with elitism and adaptive mutation is demonstrated to be near optimal as obtained with an exhaustive search . it has been demonstrated in this paper that ga achieves about 98–99 % of system sum rate as obtained with dpc with significant reduction in time and computational complexity', ['scheduling problem', 'genetic algorithm'])\n", "('method used for task', 'evolutionary algorithm ( genetic algorithm ) can be used as an alternative for this optimization problem to reduce the complexity of the search ( XXXXX ) . the performance of the genetic algorithm with elitism and adaptive mutation is demonstrated to be near optimal as obtained with an XXXXX . it has been demonstrated in this paper that ga achieves about 98–99 % of system sum rate as obtained with dpc with significant reduction in time and computational complexity', ['scheduling problem', 'exhaustive search'])\n", "('NONE', 'evolutionary algorithm ( genetic algorithm ) can be used as an alternative for this optimization problem to reduce the complexity of the search ( XXXXX ) . the performance of the genetic algorithm with elitism and adaptive mutation is demonstrated to be near optimal as obtained with an exhaustive search . it has been demonstrated in this paper that ga achieves about 98–99 % of system sum rate as obtained with dpc with significant reduction in time and XXXXX', ['scheduling problem', 'computational complexity'])\n", "('NONE', 'evolutionary algorithm ( XXXXX ) can be used as an alternative for this optimization problem to reduce the complexity of the search ( scheduling problem ) . the performance of the XXXXX with elitism and adaptive mutation is demonstrated to be near optimal as obtained with an XXXXX . it has been demonstrated in this paper that ga achieves about 98–99 % of system sum rate as obtained with dpc with significant reduction in time and computational complexity', ['genetic algorithm', 'exhaustive search'])\n", "('NONE', 'evolutionary algorithm ( XXXXX ) can be used as an alternative for this optimization problem to reduce the complexity of the search ( scheduling problem ) . the performance of the XXXXX with elitism and adaptive mutation is demonstrated to be near optimal as obtained with an exhaustive search . it has been demonstrated in this paper that ga achieves about 98–99 % of system sum rate as obtained with dpc with significant reduction in time and XXXXX', ['genetic algorithm', 'computational complexity'])\n", "('NONE', 'evolutionary algorithm ( genetic algorithm ) can be used as an alternative for this optimization problem to reduce the complexity of the search ( scheduling problem ) . the performance of the genetic algorithm with elitism and adaptive mutation is demonstrated to be near optimal as obtained with an XXXXX . it has been demonstrated in this paper that ga achieves about 98–99 % of system sum rate as obtained with dpc with significant reduction in time and XXXXX', ['exhaustive search', 'computational complexity'])\n", "('method used for task', 'novel XXXXX based on multi-gene XXXXX called gpfis-class', ['genetic fuzzy system', 'genetic programming'])\n", "('NONE', 'gpfis-class builds XXXXX premises by using multi-gene XXXXX', ['fuzzy rule', 'genetic programming'])\n", "('NONE', 'this paper address the bodyguard XXXXX ( bap ) through an XXXXX', ['allocation problem', 'evolutionary approach'])\n", "('NONE', 'we develop a XXXXX ( sa ) with special XXXXX and restart strategy for mc-top-mtw', ['simulated annealing', 'solution encoding'])\n", "('NONE', 'we develop a XXXXX ( sa ) with special solution encoding and XXXXX for mc-top-mtw', ['simulated annealing', 'restart strategy'])\n", "('method used for task', 'we develop a simulated annealing ( sa ) with special XXXXX and XXXXX for mc-top-mtw', ['solution encoding', 'restart strategy'])\n", "('NONE', 'experimental results demonstrate that mmtd has significantly performed better than state-of-the-art moeas such as moea/d , nsga-ii , amalgam , mopso : a proposal for multiple objective XXXXX , mosade , decmosa-sqp in terms of reducing the inverted generational distance ( igd ) metric and XXXXX on both zdt test problems [ 68 ] and cec’09 test instances', ['particle swarm optimization', 'time cost'])\n", "('NONE', 'to ensure good XXXXX , a XXXXX for manets must change its routing policy online to account for changes in network conditions and to deal with routing information imprecision', ['network performance', 'routing protocol'])\n", "('method used for task', 'the main focus of this paper is on the use of XXXXX and XXXXX', ['fuzzy logic', 'reinforcement learning'])\n", "('NONE', 'a fuzzy extension of a XXXXX based XXXXX for manets is presented', ['reinforcement learning', 'routing protocol'])\n", "('method used for task', 'dynamic XXXXX is more appropriate than XXXXX for adaptive energy aware routing in manets', ['fuzzy logic', 'reinforcement learning'])\n", "('method used for task', 'proposed XXXXX and variable reduction strategy reduces XXXXX', ['equality constraints', 'equality constraint'])\n", "('NONE', 'proposed equality constraint and XXXXX reduces XXXXX', ['equality constraints', 'variable reduction strategy'])\n", "('method used for task', 'proposed XXXXX and XXXXX reduces equality constraints', ['equality constraint', 'variable reduction strategy'])\n", "('method used for task', 'proposed XXXXX and XXXXX also reduces the variables of cops', ['equality constraint', 'variable reduction strategy'])\n", "('method used for task', 'a new improved XXXXX , one rank cuckoo search algorithm ( orcsa ) , is applied for solving economic XXXXX problem', ['meta-heuristic algorithm', 'load dispatch'])\n", "('method used for task', 'the advantages of the proposed orcsa are few XXXXX , high solution quality and fast XXXXX', ['control parameters', 'computational time'])\n", "('NONE', 'the advantages of the proposed orcsa are few XXXXX , high XXXXX and fast computational time', ['control parameters', 'solution quality'])\n", "('method used for task', 'the advantages of the proposed orcsa are few control parameters , high XXXXX and fast XXXXX', ['computational time', 'solution quality'])\n", "('method used for task', 'automatic XXXXX for XXXXX malignancy grading', ['clinical decision support system', 'breast cancer'])\n", "('method used for task', 'XXXXX is the most efficient XXXXX for this problem', ['particle swarm', 'search method'])\n", "('method used for task', 'we introduce three kinds of XXXXX for dual XXXXX , which the corresponding support measures can be obtained', ['distance measures', 'hesitant fuzzy sets'])\n", "('method used for task', 'we propose several power aggregation operators on dual XXXXX , study their properties and give some specific dual hesitant XXXXX', ['hesitant fuzzy sets', 'fuzzy aggregation operators'])\n", "('method used for task', 'the proposed multiple inputs and outputs ( mio ) classification method designated as the fvm-index method integrates XXXXX ( fst ) , variable precision rough set ( vprs ) theory , and a modified XXXXX ( mcvi ) function , and is designed specifically to filter out the uncertainty and inaccuracy inherent in the surveyed mio real-valued dataset ; thereby improving the classification performance', ['fuzzy set theory', 'cluster validity index'])\n", "('method used for task', 'the proposed multiple inputs and outputs ( mio ) classification method designated as the fvm-index method integrates XXXXX ( fst ) , variable precision rough set ( vprs ) theory , and a modified cluster validity index ( mcvi ) function , and is designed specifically to filter out the uncertainty and inaccuracy inherent in the surveyed mio real-valued dataset ; thereby improving the XXXXX', ['fuzzy set theory', 'classification performance'])\n", "('NONE', 'the proposed multiple inputs and outputs ( mio ) XXXXX designated as the fvm-index method integrates XXXXX ( fst ) , variable precision rough set ( vprs ) theory , and a modified cluster validity index ( mcvi ) function , and is designed specifically to filter out the uncertainty and inaccuracy inherent in the surveyed mio real-valued dataset ; thereby improving the classification performance', ['fuzzy set theory', 'classification method'])\n", "('method used for task', 'the proposed multiple inputs and outputs ( mio ) classification method designated as the fvm-index method integrates fuzzy set theory ( fst ) , variable precision rough set ( vprs ) theory , and a modified XXXXX ( mcvi ) function , and is designed specifically to filter out the uncertainty and inaccuracy inherent in the surveyed mio real-valued dataset ; thereby improving the XXXXX', ['cluster validity index', 'classification performance'])\n", "('NONE', 'the proposed multiple inputs and outputs ( mio ) XXXXX designated as the fvm-index method integrates fuzzy set theory ( fst ) , variable precision rough set ( vprs ) theory , and a modified XXXXX ( mcvi ) function , and is designed specifically to filter out the uncertainty and inaccuracy inherent in the surveyed mio real-valued dataset ; thereby improving the classification performance', ['cluster validity index', 'classification method'])\n", "('method used for task', 'the proposed multiple inputs and outputs ( mio ) XXXXX designated as the fvm-index method integrates fuzzy set theory ( fst ) , variable precision rough set ( vprs ) theory , and a modified cluster validity index ( mcvi ) function , and is designed specifically to filter out the uncertainty and inaccuracy inherent in the surveyed mio real-valued dataset ; thereby improving the XXXXX', ['classification performance', 'classification method'])\n", "('NONE', 'XXXXX leverages on XXXXX ( rl ) to enhance network security', ['cognitive radio', 'reinforcement learning'])\n", "('NONE', 'we XXXXX by considering the asset returns as XXXXX', ['model uncertainty', 'trapezoidal fuzzy numbers'])\n", "('NONE', 'we XXXXX by considering the XXXXX as trapezoidal fuzzy numbers', ['model uncertainty', 'asset returns'])\n", "('NONE', 'we model uncertainty by considering the XXXXX as XXXXX', ['trapezoidal fuzzy numbers', 'asset returns'])\n", "('NONE', \"the model is validated by applying it to a real XXXXX in ecuador 's XXXXX\", ['case study', 'ict sector'])\n", "('method used for task', 'the XXXXX is solved using a XXXXX', ['optimization problem', 'genetic algorithm'])\n", "('method used for task', 'achieved balanced clustering for enhanced XXXXX and minimum XXXXX', ['network lifetime', 'packet loss'])\n", "('method used for task', 'the XXXXX function is implemented as a XXXXX', ['neural network', 'pairwise preference'])\n", "('method used for task', 'fuzzy least squares regression ( flsr ) and XXXXX based modeling methods , switching fuzzy c-regression ( sfcr ) and takagi–sugeno ( ts ) XXXXX , were used for modeling of multi-response experiment data with replicated response measures', ['fuzzy clustering', 'fuzzy model'])\n", "('method used for task', 'fuzzy least squares regression ( flsr ) and XXXXX based modeling methods , switching fuzzy c-regression ( sfcr ) and takagi–sugeno ( ts ) fuzzy model , were used for modeling of multi-response XXXXX with replicated response measures', ['fuzzy clustering', 'experiment data'])\n", "('NONE', 'fuzzy least squares regression ( flsr ) and XXXXX based modeling methods , switching fuzzy c-regression ( sfcr ) and takagi–sugeno ( ts ) fuzzy model , were used for modeling of multi-response experiment data with replicated XXXXX', ['fuzzy clustering', 'response measures'])\n", "('NONE', 'fuzzy least squares regression ( flsr ) and fuzzy clustering based modeling methods , switching fuzzy c-regression ( sfcr ) and takagi–sugeno ( ts ) XXXXX , were used for modeling of multi-response XXXXX with replicated response measures', ['fuzzy model', 'experiment data'])\n", "('NONE', 'fuzzy least squares regression ( flsr ) and fuzzy clustering based modeling methods , switching fuzzy c-regression ( sfcr ) and takagi–sugeno ( ts ) XXXXX , were used for modeling of multi-response experiment data with replicated XXXXX', ['fuzzy model', 'response measures'])\n", "('NONE', 'fuzzy least squares regression ( flsr ) and fuzzy clustering based modeling methods , switching fuzzy c-regression ( sfcr ) and takagi–sugeno ( ts ) fuzzy model , were used for modeling of multi-response XXXXX with replicated XXXXX', ['experiment data', 'response measures'])\n", "('method used for task', 'it was seen that the sfcr had the better XXXXX rather than flsr and ts fuzzy model according to the XXXXX square error ( rmse )', ['prediction performance', 'root mean'])\n", "('method used for task', 'we introduce a suitable distance formula for XXXXX , which is compared with the XXXXX , euclidean distance , XXXXX based on hausdorff metric and signed distance', ['interval type-2 fuzzy sets', 'hamming distance'])\n", "('method used for task', 'we introduce a suitable distance formula for XXXXX , which is compared with the hamming distance , XXXXX , hamming distance based on hausdorff metric and signed distance', ['interval type-2 fuzzy sets', 'euclidean distance'])\n", "('method used for task', 'we introduce a suitable distance formula for XXXXX , which is compared with the XXXXX , euclidean distance , XXXXX based on hausdorff metric and signed distance', ['interval type-2 fuzzy sets', 'hamming distance'])\n", "('method used for task', 'we introduce a suitable XXXXX for XXXXX , which is compared with the hamming distance , euclidean distance , hamming distance based on hausdorff metric and signed distance', ['interval type-2 fuzzy sets', 'distance formula'])\n", "('NONE', 'we introduce a suitable distance formula for interval type-2 fuzzy sets , which is compared with the XXXXX , XXXXX , XXXXX based on hausdorff metric and signed distance', ['hamming distance', 'euclidean distance'])\n", "('method used for task', 'we introduce a suitable XXXXX for interval type-2 fuzzy sets , which is compared with the XXXXX , euclidean distance , XXXXX based on hausdorff metric and signed distance', ['hamming distance', 'distance formula'])\n", "('NONE', 'we introduce a suitable distance formula for interval type-2 fuzzy sets , which is compared with the XXXXX , XXXXX , XXXXX based on hausdorff metric and signed distance', ['euclidean distance', 'hamming distance'])\n", "('method used for task', 'we introduce a suitable XXXXX for interval type-2 fuzzy sets , which is compared with the hamming distance , XXXXX , hamming distance based on hausdorff metric and signed distance', ['euclidean distance', 'distance formula'])\n", "('method used for task', 'we introduce a suitable XXXXX for interval type-2 fuzzy sets , which is compared with the XXXXX , euclidean distance , XXXXX based on hausdorff metric and signed distance', ['hamming distance', 'distance formula'])\n", "('method used for task', 'based on the proposed XXXXX , we propose a hierarchical clustering-based method to solve a XXXXX and find the proximity of the suppliers', ['distance formula', 'supplier selection problem'])\n", "('method used for task', 'to illustrate the applicability of the proposed method , two examples are illustrated . the first example is a XXXXX in supplier selection problem . the next one is an example taken from the literature . then , to test the hierarchical clustering-based method and compare the obtained results by two other methods , a XXXXX using experimental analysis was designed', ['case study', 'comparative study'])\n", "('method used for task', 'to illustrate the applicability of the proposed method , two examples are illustrated . the first example is a XXXXX in supplier selection problem . the next one is an example taken from the literature . then , to test the hierarchical clustering-based method and compare the obtained results by two other methods , a comparative study using XXXXX was designed', ['case study', 'experimental analysis'])\n", "('NONE', 'to illustrate the applicability of the proposed method , two examples are illustrated . the first example is a XXXXX in XXXXX . the next one is an example taken from the literature . then , to test the hierarchical clustering-based method and compare the obtained results by two other methods , a comparative study using experimental analysis was designed', ['case study', 'supplier selection problem'])\n", "('NONE', 'to illustrate the applicability of the proposed method , two examples are illustrated . the first example is a case study in supplier selection problem . the next one is an example taken from the literature . then , to test the hierarchical clustering-based method and compare the obtained results by two other methods , a XXXXX using XXXXX was designed', ['comparative study', 'experimental analysis'])\n", "('method used for task', 'to illustrate the applicability of the proposed method , two examples are illustrated . the first example is a case study in XXXXX . the next one is an example taken from the literature . then , to test the hierarchical clustering-based method and compare the obtained results by two other methods , a XXXXX using experimental analysis was designed', ['comparative study', 'supplier selection problem'])\n", "('method used for task', 'to illustrate the applicability of the proposed method , two examples are illustrated . the first example is a case study in XXXXX . the next one is an example taken from the literature . then , to test the hierarchical clustering-based method and compare the obtained results by two other methods , a comparative study using XXXXX was designed', ['experimental analysis', 'supplier selection problem'])\n", "('method used for task', 'the novelties of our work are in both theory and application . we propose a new distance formula for XXXXX . based on the proposed distance formula , we propose a new XXXXX to solve a supplier selection problem and finally to illustrate the applicability of the proposed method , a case study is used', ['interval type-2 fuzzy sets', 'clustering method'])\n", "('method used for task', 'the novelties of our work are in both theory and application . we propose a new distance formula for XXXXX . based on the proposed distance formula , we propose a new clustering method to solve a supplier selection problem and finally to illustrate the applicability of the proposed method , a XXXXX is used', ['interval type-2 fuzzy sets', 'case study'])\n", "('method used for task', 'the novelties of our work are in both theory and application . we propose a new XXXXX for XXXXX . based on the proposed XXXXX , we propose a new clustering method to solve a supplier selection problem and finally to illustrate the applicability of the proposed method , a case study is used', ['interval type-2 fuzzy sets', 'distance formula'])\n", "('method used for task', 'the novelties of our work are in both theory and application . we propose a new XXXXX for XXXXX . based on the proposed XXXXX , we propose a new clustering method to solve a supplier selection problem and finally to illustrate the applicability of the proposed method , a case study is used', ['interval type-2 fuzzy sets', 'distance formula'])\n", "('method used for task', 'the novelties of our work are in both theory and application . we propose a new distance formula for XXXXX . based on the proposed distance formula , we propose a new clustering method to solve a XXXXX and finally to illustrate the applicability of the proposed method , a case study is used', ['interval type-2 fuzzy sets', 'supplier selection problem'])\n", "('method used for task', 'the novelties of our work are in both theory and application . we propose a new distance formula for interval type-2 fuzzy sets . based on the proposed distance formula , we propose a new XXXXX to solve a supplier selection problem and finally to illustrate the applicability of the proposed method , a XXXXX is used', ['clustering method', 'case study'])\n", "('method used for task', 'the novelties of our work are in both theory and application . we propose a new XXXXX for interval type-2 fuzzy sets . based on the proposed XXXXX , we propose a new XXXXX to solve a supplier selection problem and finally to illustrate the applicability of the proposed method , a case study is used', ['clustering method', 'distance formula'])\n", "('method used for task', 'the novelties of our work are in both theory and application . we propose a new XXXXX for interval type-2 fuzzy sets . based on the proposed XXXXX , we propose a new XXXXX to solve a supplier selection problem and finally to illustrate the applicability of the proposed method , a case study is used', ['clustering method', 'distance formula'])\n", "('method used for task', 'the novelties of our work are in both theory and application . we propose a new distance formula for interval type-2 fuzzy sets . based on the proposed distance formula , we propose a new XXXXX to solve a XXXXX and finally to illustrate the applicability of the proposed method , a case study is used', ['clustering method', 'supplier selection problem'])\n", "('method used for task', 'the novelties of our work are in both theory and application . we propose a new XXXXX for interval type-2 fuzzy sets . based on the proposed XXXXX , we propose a new clustering method to solve a supplier selection problem and finally to illustrate the applicability of the proposed method , a XXXXX is used', ['case study', 'distance formula'])\n", "('method used for task', 'the novelties of our work are in both theory and application . we propose a new XXXXX for interval type-2 fuzzy sets . based on the proposed XXXXX , we propose a new clustering method to solve a supplier selection problem and finally to illustrate the applicability of the proposed method , a XXXXX is used', ['case study', 'distance formula'])\n", "('NONE', 'the novelties of our work are in both theory and application . we propose a new distance formula for interval type-2 fuzzy sets . based on the proposed distance formula , we propose a new clustering method to solve a XXXXX and finally to illustrate the applicability of the proposed method , a XXXXX is used', ['case study', 'supplier selection problem'])\n", "('method used for task', 'the novelties of our work are in both theory and application . we propose a new XXXXX for interval type-2 fuzzy sets . based on the proposed XXXXX , we propose a new clustering method to solve a XXXXX and finally to illustrate the applicability of the proposed method , a case study is used', ['distance formula', 'supplier selection problem'])\n", "('method used for task', 'the novelties of our work are in both theory and application . we propose a new XXXXX for interval type-2 fuzzy sets . based on the proposed XXXXX , we propose a new clustering method to solve a XXXXX and finally to illustrate the applicability of the proposed method , a case study is used', ['distance formula', 'supplier selection problem'])\n", "('method used for task', 'large scale problems require large XXXXX and XXXXX', ['population size', 'computational cost'])\n", "('NONE', 'XXXXX require large XXXXX and computational cost', ['population size', 'large scale problems'])\n", "('method used for task', 'XXXXX require large population size and XXXXX', ['computational cost', 'large scale problems'])\n", "('NONE', 'new justification of the XXXXX by XXXXX', ['theoretical model', 'hesitant fuzzy sets'])\n", "('NONE', 'it decreases the uncertainty and the XXXXX in XXXXX', ['information loss', 'group decision making'])\n", "('method used for task', 'further study can focus on interval type 2 XXXXX to develop a XXXXX', ['fuzzy set', 'decision support system'])\n", "('NONE', 'the interpretability measure is based on the XXXXX of three components : the average length of rules , the number of active XXXXX , and the number of active inputs of the system', ['arithmetic mean', 'fuzzy sets'])\n", "('NONE', 'a novel method for XXXXX with intuitionistic XXXXX is proposed and applied to radio frequency identification ( rfid ) technology selection', ['group decision making', 'fuzzy preference relations'])\n", "('method used for task', 'a novel method for XXXXX with intuitionistic fuzzy preference relations is proposed and applied to radio frequency identification ( rfid ) XXXXX', ['group decision making', 'technology selection'])\n", "('method used for task', 'a novel method for group decision making with intuitionistic XXXXX is proposed and applied to radio frequency identification ( rfid ) XXXXX', ['fuzzy preference relations', 'technology selection'])\n", "('NONE', 'we introduce a XXXXX considering articles on XXXXX in business published in last two decades', ['literature review', 'artificial neural networks'])\n", "('NONE', 'a synthetic XXXXX using XXXXX is presented', ['evaluation method', 'triangular fuzzy number'])\n", "('NONE', 'we get a speech signal model on a XXXXX by XXXXX', ['reconstructed phase space', 'chaotic time series'])\n", "('method used for task', 'presents an XXXXX for finding XXXXX in different dimensions for each particle', ['adaptive method', 'inertia weight'])\n", "('method used for task', 'XXXXX is employed to deal with fuzziness , and a set of XXXXX is obtained', ['ranking function', 'efficient solutions'])\n", "('NONE', 'proposed a system for grammar inference by utilizing the mask-fill XXXXX and boolean based procedure with minimum XXXXX principle', ['reproduction operators', 'description length'])\n", "('NONE', 'we design a new XXXXX which takes into account two term , the XXXXX and the class separability distance', ['objective function', 'classification error rate'])\n", "('method used for task', 'the framework also considers XXXXX and XXXXX in dea models', ['high-dimensional data', 'missing values'])\n", "('NONE', 'the framework also considers XXXXX and missing values in XXXXX', ['high-dimensional data', 'dea models'])\n", "('NONE', 'the framework also considers high-dimensional data and XXXXX in XXXXX', ['missing values', 'dea models'])\n", "('NONE', 'a dimension-reduction method improves the XXXXX of the XXXXX', ['discrimination power', 'dea model'])\n", "('NONE', 'a preference ratio method ranks the interval XXXXX in the XXXXX', ['fuzzy environment', 'efficiency scores'])\n", "('NONE', 'equipping the svr model with ann as XXXXX has improved the XXXXX', ['kernel function', 'model accuracy'])\n", "('method used for task', 'a maximizing XXXXX is established for determining XXXXX', ['optimization model', 'criteria weights'])\n", "('method used for task', 'an experimental hardware , including the proposed XXXXX method is set up . the experimental results have confirmed that the XXXXX method exhibits satisfactory predicting performance for svi', ['soft computing', 'soft computing method'])\n", "('method used for task', 'an experimental hardware , including the proposed XXXXX method is set up . the experimental results have confirmed that the XXXXX method exhibits satisfactory predicting performance for svi', ['soft computing', 'soft computing method'])\n", "('method used for task', 'XXXXX is the graphical analysis by which the behavior between certain XXXXX of the antenna can be visually and numerically analyzed', ['curve fitting', 'design parameters'])\n", "('method used for task', 'XXXXX is the individual element called the particles form a swarm , all particles attains one initial position randomly at the start and then they fly through the genotypic space to find the XXXXX of the problem', ['particle swarm optimization', 'optimal solution'])\n", "('method used for task', 'we present a study on the adaptation of the island sizes of a distributed XXXXX to the XXXXX of the nodes of an heterogeneous cluster', ['evolutionary algorithm', 'computational power'])\n", "('NONE', 'firstly , the XXXXX uses XXXXX ( ga ) with a high exploration power', ['hybrid method', 'genetic algorithm'])\n", "('NONE', 'detail flow diagram of processes for obtaining the proposed solution of fp equation using XXXXX trained with XXXXX ga-sqp', ['neural networks', 'hybrid approach'])\n", "('NONE', 'XXXXX has been solved using XXXXX', ['optimization problem', 'genetic algorithms'])\n", "('method used for task', 'the XXXXX is equipped with a mixed XXXXX for the gait learning', ['kernel function', 'svm controller'])\n", "('method used for task', 'the XXXXX is trained based on XXXXX sizes', ['small sample', 'svm controller'])\n", "('method used for task', 'a novel XXXXX based algorithm inspired by superposition principle and field attraction for XXXXX', ['swarm intelligence', 'global optimization'])\n", "('method used for task', 'the XXXXX and XXXXX of the proposed method are increased by elimination of redundant features by using different feature selection methods', ['generalization capability', 'detection accuracy'])\n", "('NONE', 'the XXXXX and detection accuracy of the proposed method are increased by elimination of redundant features by using different XXXXX', ['generalization capability', 'feature selection methods'])\n", "('NONE', 'the generalization capability and XXXXX of the proposed method are increased by elimination of redundant features by using different XXXXX', ['detection accuracy', 'feature selection methods'])\n", "('NONE', 'our evolutionary XXXXX provides improved XXXXX', ['recognition rate', 'face recognition algorithm'])\n", "('NONE', 'XXXXX of XXXXX has been assessed', ['sensitivity analysis', 'decision maker preferences'])\n", "('method used for task', 'the balance factor is designed to maintain the diversity of solutions . the XXXXX is utilized to enhance the searching ability in entire XXXXX', ['scale factor', 'solution space'])\n", "('method used for task', 'the proposed algorithm is efficient for solving the large-scale XXXXX , especially for problems that have magnitude difference between two XXXXX', ['scheduling problems', 'optimization objectives'])\n", "('method used for task', 'this paper proposed a binary version of the XXXXX for solving 0-1 XXXXX', ['knapsack problem', 'monkey algorithm'])\n", "('method used for task', 'the ant system algorithm and XXXXX were combined to solve 3d/2d fixed-outline XXXXX', ['simulated annealing', 'floorplanning problem'])\n", "('NONE', 'a novel XXXXX using multiple XXXXX is designed', ['classification model', 'base classifiers'])\n", "('NONE', 'iapso outperforms several recent XXXXX , in terms of accuracy and XXXXX', ['meta-heuristic algorithms', 'convergence speed'])\n", "('method used for task', 'the concepts of XXXXX , route planning , and XXXXX are combined', ['fuzzy modeling', 'parallel machine scheduling'])\n", "('method used for task', 'the lower level problem is modeled as a mmbm problem , and a hungarian-based XXXXX ( hlp ) method is proposed , which can solve mmbm in XXXXX', ['linear programming', 'polynomial time'])\n", "('method used for task', 'the lower XXXXX is modeled as a mmbm problem , and a hungarian-based XXXXX ( hlp ) method is proposed , which can solve mmbm in polynomial time', ['linear programming', 'level problem'])\n", "('method used for task', 'the lower XXXXX is modeled as a mmbm problem , and a hungarian-based linear programming ( hlp ) method is proposed , which can solve mmbm in XXXXX', ['polynomial time', 'level problem'])\n", "('method used for task', 'the upper level optimization problem is solved by evolutionary multiobjective optimization algorithms , where a greedy reassignment local search operator , capable of leveraging the XXXXX and information from XXXXX , is introduced to improve the efficiency of the algorithm', ['domain knowledge', 'problem instances'])\n", "('method used for task', 'an XXXXX is determined for segmentation of XXXXX', ['adaptive threshold', 'power quality signals'])\n", "('method used for task', 'the XXXXX are based on XXXXX and acquired in field', ['mathematical models', 'power quality signals'])\n", "('method used for task', 'the intersections between the XXXXX and the XXXXX detail curves determine the start and the end of the segments', ['adaptive threshold', 'wavelet transform'])\n", "('method used for task', 'we develop a XXXXX to solve the green XXXXX', ['vehicle routing problem', 'solution approach'])\n", "('NONE', 'the proposed system utilized XXXXX for indicating the degree of XXXXX', ['case-based reasoning', 'water quality'])\n", "('NONE', 'three XXXXX , four XXXXX', ['evolutionary algorithms', 'fitness functions'])\n", "('NONE', 'cost minimized in a mix-product three echelon XXXXX while maintaining XXXXX', ['supply chain', 'service level'])\n", "('NONE', 'XXXXX , XXXXX , revised simos procedure , analytical hierarchy process', ['linear regression', 'factor analysis'])\n", "('NONE', 'XXXXX , factor analysis , revised simos procedure , analytical XXXXX', ['linear regression', 'hierarchy process'])\n", "('NONE', 'linear regression , XXXXX , revised simos procedure , analytical XXXXX', ['factor analysis', 'hierarchy process'])\n", "('NONE', 'minimization of maximum XXXXX of all parts is considered as XXXXX', ['completion time', 'objective function'])\n", "('method used for task', 'the aggregated XXXXX was used to investigate the simultaneous effects of printing parameters on the XXXXX and porosity of scaffolds', ['artificial neural network', 'compressive strength'])\n", "('NONE', 'most existing approaches reduce the XXXXX of qos-aware XXXXX to a single-objective problem using scalarization', ['multi-objective optimization problem', 'web service composition'])\n", "('method used for task', 'XXXXX ( de ) algorithms – in particular gde3 – yields the best results on this specific problem for several scenarios , also having the lowest XXXXX', ['differential evolution', 'time complexity'])\n", "('method used for task', 'the XXXXX for XXXXX is a critical decision making problem', ['location selection', 'nuclear power plant'])\n", "('NONE', 'the information of the XXXXX is used for online tuning of feedback gains of the XXXXX', ['pid controllers', 'wavelet fuzzy neural network'])\n", "('method used for task', 'the XXXXX approach is used to optimize local and global modelling capability of XXXXX', ['t-s fuzzy model', 'weighting parameters'])\n", "('NONE', 'the weighting parameters approach is used to optimize local and global XXXXX of XXXXX', ['t-s fuzzy model', 'modelling capability'])\n", "('method used for task', 'the XXXXX approach is used to optimize local and global XXXXX of t-s fuzzy model', ['weighting parameters', 'modelling capability'])\n", "('method used for task', 'a new XXXXX is developed for estimating the process change point in x¯ XXXXX', ['hybrid method', 'control chart'])\n", "('method used for task', 'a new XXXXX is developed for estimating the XXXXX in x¯ control chart', ['hybrid method', 'process change point'])\n", "('NONE', 'a new hybrid method is developed for estimating the XXXXX in x¯ XXXXX', ['control chart', 'process change point'])\n", "('NONE', 'the proposed XXXXX provides a more accurate estimate of the XXXXX', ['hybrid method', 'process change point'])\n", "('NONE', 'alc-pso is applied for the solution of XXXXX of XXXXX', ['power systems', 'opf problem'])\n", "('method used for task', 'a novel method for combination of XXXXX and fuzzy XXXXX ( frl ) is proposed', ['supervised learning', 'reinforcement learning'])\n", "('NONE', 'XXXXX is used for initialization of value ( worth ) of each candidate action of XXXXX in critic-only based frl algorithms', ['supervised learning', 'fuzzy rules'])\n", "('NONE', 'since the mamdani engines perform XXXXX on disjoint subsets of the XXXXX , the total number of the fuzzy rules needed for input–output mapping is far smaller', ['fuzzy reasoning', 'linguistic variables'])\n", "('NONE', 'since the mamdani engines perform XXXXX on disjoint subsets of the linguistic variables , the total number of the XXXXX needed for input–output mapping is far smaller', ['fuzzy reasoning', 'fuzzy rules'])\n", "('NONE', 'since the mamdani engines perform fuzzy reasoning on disjoint subsets of the XXXXX , the total number of the XXXXX needed for input–output mapping is far smaller', ['linguistic variables', 'fuzzy rules'])\n", "('method used for task', 'coupling the chain classification with greedy breadth-first-search ( gbfs ) algorithm , a new approach has been proposed for automated sizing of defects , utilizing XXXXX ( rbfnn ) and XXXXX ( svm ) . it has been established that the proposed approach can successfully size subsurface as well as surface defects', ['radial basis function neural network', 'support vector machine'])\n", "('NONE', 'coupling the chain classification with greedy breadth-first-search ( gbfs ) algorithm , a new approach has been proposed for automated sizing of defects , utilizing XXXXX ( rbfnn ) and support vector machine ( svm ) . it has been established that the proposed approach can successfully size subsurface as well as XXXXX', ['radial basis function neural network', 'surface defects'])\n", "('NONE', 'coupling the XXXXX with greedy breadth-first-search ( gbfs ) algorithm , a new approach has been proposed for automated sizing of defects , utilizing XXXXX ( rbfnn ) and support vector machine ( svm ) . it has been established that the proposed approach can successfully size subsurface as well as surface defects', ['radial basis function neural network', 'chain classification'])\n", "('NONE', 'coupling the chain classification with greedy breadth-first-search ( gbfs ) algorithm , a new approach has been proposed for automated sizing of defects , utilizing radial basis function neural network ( rbfnn ) and XXXXX ( svm ) . it has been established that the proposed approach can successfully size subsurface as well as XXXXX', ['support vector machine', 'surface defects'])\n", "('method used for task', 'coupling the XXXXX with greedy breadth-first-search ( gbfs ) algorithm , a new approach has been proposed for automated sizing of defects , utilizing radial basis function neural network ( rbfnn ) and XXXXX ( svm ) . it has been established that the proposed approach can successfully size subsurface as well as surface defects', ['support vector machine', 'chain classification'])\n", "('NONE', 'coupling the XXXXX with greedy breadth-first-search ( gbfs ) algorithm , a new approach has been proposed for automated sizing of defects , utilizing radial basis function neural network ( rbfnn ) and support vector machine ( svm ) . it has been established that the proposed approach can successfully size subsurface as well as XXXXX', ['surface defects', 'chain classification'])\n", "('NONE', 'XXXXX has incorporated dependency among the XXXXX ( length , width , depth and height ) and optimal sequence for sizing has been determined , for the first time', ['chain classification', 'class variables'])\n", "('method used for task', 'XXXXX has been able to significantly incorporate the dependency existing among the XXXXX and has successfully resulted in sizing of defects located even 3mm below the surface', ['chain classification', 'class variables'])\n", "('method used for task', 'odg problem is studied with an objective of reducing XXXXX and XXXXX', ['power loss', 'energy cost'])\n", "('NONE', 'eligible cluster heads are selected in the networks using XXXXX which ensures considerable XXXXX in the network', ['fuzzy logic', 'energy saving'])\n", "('method used for task', 'the nodes having highest number of neighbors , higher XXXXX and close to XXXXX are given preference in cluster head election', ['residual energy', 'base station'])\n", "('method used for task', 'the proposed method is applied on uci XXXXX repository datasets and its corresponding XXXXX are obtained', ['machine learning', 'performance measures'])\n", "('NONE', 'the comparison of the proposed XXXXX with the existing results reveal that the accuracy obtained by the present technique provides better XXXXX', ['similarity measures', 'similarity rating'])\n", "('method used for task', 'it is also observed that fixing the XXXXX for the XXXXX is difficult due to the negative rating in some existing methodology', ['similarity measures', 'threshold values'])\n", "('method used for task', 'the XXXXX obtained by the suggested technique is compared with the XXXXX data set and are analyzed', ['similarity rating', 'class distribution'])\n", "('method used for task', 'dynamic XXXXX model for obtaining the candidate XXXXX', ['expectation efficiency', 'objective function value'])\n", "('method used for task', 'static XXXXX model for updating the XXXXX', ['expectation efficiency', 'objective function value'])\n", "('NONE', 'simultaneous estimation of XXXXX and pixel labels of XXXXX is focus of work', ['bias field', 'mr images'])\n", "('method used for task', 'constrained XXXXX is used to tune the parameters of a robust XXXXX', ['particle swarm optimization algorithm', 'pid controller'])\n", "('NONE', 'XXXXX within the intervals of the XXXXX is ensured', ['robust stability', 'uncertain parameters'])\n", "('NONE', 'the method provides both a dynamic detection and fuzzy querying of XXXXX in crime-related XXXXX', ['time series', 'change points'])\n", "('NONE', 'XXXXX expressed with XXXXX can be used to query change points', ['geometric properties', 'linguistic variables'])\n", "('method used for task', 'XXXXX expressed with linguistic variables can be used to query XXXXX', ['geometric properties', 'change points'])\n", "('method used for task', 'geometric properties expressed with XXXXX can be used to query XXXXX', ['linguistic variables', 'change points'])\n", "('method used for task', 'we study a XXXXX system for outsourcing XXXXX ( orl )', ['reverse logistics', 'criteria evaluation'])\n", "('method used for task', 'analytic XXXXX is used to find the best XXXXX', ['compromise solution', 'hierarchy process'])\n", "('method used for task', 'this paper presents a new XXXXX for XXXXX with coa', ['data clustering', 'optimization approach'])\n", "('NONE', 'this paper presents a new optimization approach for XXXXX with coa and XXXXX for clustering data', ['data clustering', 'fuzzy system'])\n", "('method used for task', 'this paper presents a new XXXXX for XXXXX with coa and fuzzy system for clustering data', ['data clustering', 'optimization approach'])\n", "('method used for task', 'this paper presents a new XXXXX for data clustering with coa and XXXXX for clustering data', ['fuzzy system', 'optimization approach'])\n", "('NONE', 'an efficient proposal for XXXXX by cuckoo XXXXX', ['data clustering', 'optimization algorithm'])\n", "('NONE', 'all XXXXX are implemented using ds1104 dsp-based XXXXX', ['control algorithms', 'control computer'])\n", "('NONE', 'selection method based on a XXXXX optimized by a XXXXX', ['fuzzy system', 'genetic algorithm'])\n", "('method used for task', 'XXXXX based on a XXXXX optimized by a genetic algorithm', ['fuzzy system', 'selection method'])\n", "('method used for task', 'XXXXX based on a fuzzy system optimized by a XXXXX', ['genetic algorithm', 'selection method'])\n", "('NONE', 'the XXXXX of the XXXXX is designed with another fuzzy system in order to accomplish it optimization task', ['fitness function', 'genetic algorithm'])\n", "('NONE', 'the XXXXX of the genetic algorithm is designed with another XXXXX in order to accomplish it optimization task', ['fitness function', 'fuzzy system'])\n", "('method used for task', 'the fitness function of the XXXXX is designed with another XXXXX in order to accomplish it optimization task', ['genetic algorithm', 'fuzzy system'])\n", "('NONE', 'the neuro-fuzzy XXXXX uses sugeno type fuzzy rules with a randomized fuzzy layer and a linear XXXXX output layer', ['inference system', 'neural network'])\n", "('NONE', 'a novel XXXXX that makes a lot of sense when one has an absolute maximum number of XXXXX allowed in your application', ['performance evaluation method', 'function evaluations'])\n", "('method used for task', 'the XXXXX and precision of conventional pso is improved by introducing an adaptive inertia weight strategy based on the XXXXX of the particles', ['convergence speed', 'success rate'])\n", "('method used for task', 'a novel XXXXX for XXXXX under uncertainty is presented', ['soft computing approach', 'location decision'])\n", "('method used for task', 'the XXXXX is introduced for the parameter hmcr to improve the performance of the proposed XXXXX', ['dynamic adaptation', 'routing algorithm'])\n", "('NONE', 'an effective local search strategy is proposed to improve the XXXXX and the accuracy of the proposed XXXXX', ['convergence speed', 'routing algorithm'])\n", "('method used for task', 'an effective XXXXX is proposed to improve the XXXXX and the accuracy of the proposed routing algorithm', ['convergence speed', 'local search strategy'])\n", "('method used for task', 'an effective XXXXX is proposed to improve the convergence speed and the accuracy of the proposed XXXXX', ['routing algorithm', 'local search strategy'])\n", "('method used for task', 'XXXXX is trained to predict the XXXXX of composites from experimental variables', ['neural network', 'compressive strength'])\n", "('NONE', 'input variables are optimized to maximize XXXXX , by XXXXX', ['compressive strength', 'genetic algorithm'])\n", "('method used for task', 'XXXXX are optimized to maximize XXXXX , by genetic algorithm', ['compressive strength', 'input variables'])\n", "('method used for task', 'XXXXX are optimized to maximize compressive strength , by XXXXX', ['genetic algorithm', 'input variables'])\n", "('NONE', 'a robust neural-network-method applied to XXXXX in line-connected three-phase XXXXX is presented', ['speed estimation', 'induction motors'])\n", "('method used for task', 'a XXXXX of single and multiple current sensor applied to XXXXX is presented', ['comparative study', 'speed estimation'])\n", "('method used for task', 'we propose multiobjective XXXXX for XXXXX', ['evolutionary computation', 'multiple sequence alignment'])\n", "('NONE', 'unrelated parallel machines scheduling problem with past-sequence-dependent XXXXX , XXXXX , deteriorating jobs and learning effects is presented', ['setup times', 'release dates'])\n", "('NONE', 'unrelated parallel machines scheduling problem with past-sequence-dependent XXXXX , release dates , XXXXX and learning effects is presented', ['setup times', 'deteriorating jobs'])\n", "('method used for task', 'unrelated parallel machines scheduling problem with past-sequence-dependent XXXXX , release dates , deteriorating jobs and XXXXX is presented', ['setup times', 'learning effects'])\n", "('NONE', 'unrelated parallel machines scheduling problem with past-sequence-dependent setup times , XXXXX , XXXXX and learning effects is presented', ['release dates', 'deteriorating jobs'])\n", "('method used for task', 'unrelated parallel machines scheduling problem with past-sequence-dependent setup times , XXXXX , deteriorating jobs and XXXXX is presented', ['release dates', 'learning effects'])\n", "('method used for task', 'unrelated parallel machines scheduling problem with past-sequence-dependent setup times , release dates , XXXXX and XXXXX is presented', ['deteriorating jobs', 'learning effects'])\n", "('method used for task', 'XXXXX takes least estimation time and surpasses all other 11 XXXXX', ['firefly algorithm', 'metaheuristic algorithms'])\n", "('method used for task', 'results are compared with the results obtained using XXXXX and artificial bee XXXXX and it is proved that the proposed method offers reduced thd with less computation period', ['particle swarm optimization', 'colony algorithm'])\n", "('NONE', 'we studied the effect of XXXXX on multi-objective XXXXX', ['local search', 'memetic algorithm'])\n", "('NONE', 'morris’ oat method drives the XXXXX of the XXXXX', ['neighborhood search', 'abc algorithm'])\n", "('method used for task', \"the XXXXX is developed in nvidia 's cuda/c to reduce the XXXXX\", ['execution time', 'identification algorithm'])\n", "('method used for task', 'pfs models are compared with takagi-sugeno XXXXX and XXXXX', ['fuzzy models', 'logistic regression models'])\n", "('NONE', 'the application of using XXXXX in the XXXXX can be adopted suitably . all of members were able to assess dependently without any bias from the team members', ['fuzzy fmea', 'emergency department'])\n", "('method used for task', 'XXXXX can be applied for the first time to improve XXXXX in an emergency department of a public hospital', ['fuzzy fmea', 'decision making process'])\n", "('method used for task', 'XXXXX can be applied for the first time to improve decision making process in an XXXXX of a public hospital', ['fuzzy fmea', 'emergency department'])\n", "('NONE', 'fuzzy fmea can be applied for the first time to improve XXXXX in an XXXXX of a public hospital', ['decision making process', 'emergency department'])\n", "('NONE', 'the XXXXX are tuned by XXXXX', ['control parameters', 'fuzzy logic rules'])\n", "('NONE', 'the XXXXX was confirmed by XXXXX', ['performance improvement', 'simulation study'])\n", "('method used for task', 'XXXXX are carried out for XXXXX', ['computer simulations', 'performance evaluation'])\n", "('method used for task', 'conducting the XXXXX to understand in a better way the XXXXX', ['case study', 'evaluation process'])\n", "('method used for task', 'a XXXXX based on XXXXX is presented', ['binary particle swarm optimization', 'feature selection method'])\n", "('method used for task', 'fitness based adaptive XXXXX is integrated with the XXXXX to dynamically control the exploration and exploitation of the particle in the search space', ['inertia weight', 'binary particle swarm optimization'])\n", "('method used for task', 'fitness based adaptive XXXXX is integrated with the binary particle swarm optimization to dynamically control the exploration and exploitation of the particle in the XXXXX', ['inertia weight', 'search space'])\n", "('NONE', 'fitness based adaptive inertia weight is integrated with the XXXXX to dynamically control the exploration and exploitation of the particle in the XXXXX', ['binary particle swarm optimization', 'search space'])\n", "('method used for task', 'the kernel principal component analysis is used to transform nonlinear XXXXX to improve XXXXX', ['feature space', 'classification performance'])\n", "('method used for task', 'XXXXX , which is regarded as an important and practically useful XXXXX , is fully exploited for designing sugeno-type granular model', ['information granularity', 'design asset'])\n", "('NONE', 'the proposed approach of constructing the sugeno-type granular model is of general nature as it could be applied to various XXXXX and realized by invoking different formalisms of XXXXX', ['fuzzy models', 'information granules'])\n", "('method used for task', 'a XXXXX based on hesitant fuzzy XXXXX is proposed.it is capable of taking into account the fundamental building blocks of trust.it considers the hesitancy and vagueness of trust decision making process.it considers the subjectivity and uncertainty of trust decision making process.its effectiveness is demonstrated by providing several evaluation scenarios', ['trust model', 'multi-criteria decision making'])\n", "('method used for task', 'a XXXXX based on hesitant fuzzy multi-criteria decision making is proposed.it is capable of taking into account the fundamental XXXXX of trust.it considers the hesitancy and vagueness of trust decision making process.it considers the subjectivity and uncertainty of trust decision making process.its effectiveness is demonstrated by providing several evaluation scenarios', ['trust model', 'building blocks'])\n", "('method used for task', 'a XXXXX based on hesitant fuzzy multi-criteria XXXXX is proposed.it is capable of taking into account the fundamental building blocks of trust.it considers the hesitancy and vagueness of trust XXXXX process.it considers the subjectivity and uncertainty of trust XXXXX process.its effectiveness is demonstrated by providing several evaluation scenarios', ['trust model', 'decision making'])\n", "('method used for task', 'a XXXXX based on hesitant fuzzy multi-criteria decision making is proposed.it is capable of taking into account the fundamental building blocks of trust.it considers the hesitancy and vagueness of XXXXX process.it considers the subjectivity and uncertainty of XXXXX process.its effectiveness is demonstrated by providing several evaluation scenarios', ['trust model', 'trust decision making'])\n", "('method used for task', 'a XXXXX based on hesitant fuzzy multi-criteria decision making is proposed.it is capable of taking into account the fundamental building blocks of trust.it considers the hesitancy and vagueness of XXXXX process.it considers the subjectivity and uncertainty of XXXXX process.its effectiveness is demonstrated by providing several evaluation scenarios', ['trust model', 'trust decision making'])\n", "('method used for task', 'a trust model based on hesitant fuzzy XXXXX is proposed.it is capable of taking into account the fundamental XXXXX of trust.it considers the hesitancy and vagueness of trust decision making process.it considers the subjectivity and uncertainty of trust decision making process.its effectiveness is demonstrated by providing several evaluation scenarios', ['multi-criteria decision making', 'building blocks'])\n", "('NONE', 'a trust model based on hesitant fuzzy XXXXX is proposed.it is capable of taking into account the fundamental building blocks of trust.it considers the hesitancy and vagueness of trust XXXXX process.it considers the subjectivity and uncertainty of trust XXXXX process.its effectiveness is demonstrated by providing several evaluation scenarios', ['multi-criteria decision making', 'decision making'])\n", "('NONE', 'a trust model based on hesitant fuzzy XXXXX is proposed.it is capable of taking into account the fundamental building blocks of trust.it considers the hesitancy and vagueness of XXXXX process.it considers the subjectivity and uncertainty of XXXXX process.its effectiveness is demonstrated by providing several evaluation scenarios', ['multi-criteria decision making', 'trust decision making'])\n", "('NONE', 'a trust model based on hesitant fuzzy XXXXX is proposed.it is capable of taking into account the fundamental building blocks of trust.it considers the hesitancy and vagueness of XXXXX process.it considers the subjectivity and uncertainty of XXXXX process.its effectiveness is demonstrated by providing several evaluation scenarios', ['multi-criteria decision making', 'trust decision making'])\n", "('method used for task', 'a trust model based on hesitant fuzzy multi-criteria XXXXX is proposed.it is capable of taking into account the fundamental XXXXX of trust.it considers the hesitancy and vagueness of trust XXXXX process.it considers the subjectivity and uncertainty of trust XXXXX process.its effectiveness is demonstrated by providing several evaluation scenarios', ['building blocks', 'decision making'])\n", "('NONE', 'a trust model based on hesitant fuzzy multi-criteria decision making is proposed.it is capable of taking into account the fundamental XXXXX of trust.it considers the hesitancy and vagueness of XXXXX process.it considers the subjectivity and uncertainty of XXXXX process.its effectiveness is demonstrated by providing several evaluation scenarios', ['building blocks', 'trust decision making'])\n", "('NONE', 'a trust model based on hesitant fuzzy multi-criteria decision making is proposed.it is capable of taking into account the fundamental XXXXX of trust.it considers the hesitancy and vagueness of XXXXX process.it considers the subjectivity and uncertainty of XXXXX process.its effectiveness is demonstrated by providing several evaluation scenarios', ['building blocks', 'trust decision making'])\n", "('NONE', 'a trust model based on hesitant fuzzy multi-criteria XXXXX is proposed.it is capable of taking into account the fundamental building blocks of trust.it considers the hesitancy and vagueness of trust XXXXX process.it considers the subjectivity and uncertainty of trust XXXXX process.its effectiveness is demonstrated by providing several evaluation scenarios', ['decision making', 'trust decision making'])\n", "('NONE', 'a trust model based on hesitant fuzzy multi-criteria XXXXX is proposed.it is capable of taking into account the fundamental building blocks of trust.it considers the hesitancy and vagueness of trust XXXXX process.it considers the subjectivity and uncertainty of trust XXXXX process.its effectiveness is demonstrated by providing several evaluation scenarios', ['decision making', 'trust decision making'])\n", "('NONE', 'XXXXX shows that XXXXX is efficient to work with gmf', ['comparative analysis', 'differential evolution'])\n", "('NONE', 'we propose a XXXXX by propagating probabilities between XXXXX', ['clustering algorithm', 'data points'])\n", "('method used for task', 'develop new XXXXX for interval reciprocal XXXXX ( irprs )', ['similarity measures', 'preference relations'])\n", "('method used for task', 'a hybrid XXXXX based on XXXXX is proposed', ['particle swarm optimization', 'feature selection method'])\n", "('method used for task', 'our method uses a novel XXXXX to enhance the XXXXX near global optima', ['local search', 'search process'])\n", "('method used for task', 'natural XXXXX is used in XXXXX', ['evolution strategy', 'structural optimization'])\n", "('method used for task', 'using XXXXX for XXXXX and rule combination', ['genetic algorithm', 'parameter selection'])\n", "('NONE', 'applying hybrid XXXXX on real life very XXXXX', ['evolutionary methods', 'large data set'])\n", "('method used for task', 'to reduce the XXXXX , the modified cnfslms algorithm is proposed by replacing the XXXXX with a versorial function', ['computational complexity', 'sigmoid function'])\n", "('NONE', 'the idea treats XXXXX as an essential XXXXX', ['information granularity', 'design asset'])\n", "('method used for task', 'XXXXX based on XXXXX are completed', ['comparative studies', 'global sensitivity analysis'])\n", "('method used for task', 'parallel XXXXX ( pvns ) is implemented for solving bus terminal XXXXX ( btlp )', ['variable neighborhood search', 'location problem'])\n", "('method used for task', 'improved XXXXX based on the neighborhoods fast interchange is combined with reduced XXXXX based on the covering characteristic of the problem', ['local search', 'neighborhood size'])\n", "('NONE', 'we study the XXXXX of finding a XXXXX of graphs based on rough sets', ['np-hard problem', 'minimum vertex cover'])\n", "('NONE', 'we study the XXXXX of finding a minimum vertex cover of graphs based on XXXXX', ['np-hard problem', 'rough sets'])\n", "('NONE', 'we study the np-hard problem of finding a XXXXX of graphs based on XXXXX', ['minimum vertex cover', 'rough sets'])\n", "('NONE', 'the problem of finding a XXXXX of graphs can be translated into the problem of finding an optimal reduct of a decision information table in XXXXX', ['minimum vertex cover', 'rough sets'])\n", "('NONE', 'the problem of finding a XXXXX of graphs can be translated into the problem of finding an optimal reduct of a XXXXX table in rough sets', ['minimum vertex cover', 'decision information'])\n", "('NONE', 'the problem of finding a minimum vertex cover of graphs can be translated into the problem of finding an optimal reduct of a XXXXX table in XXXXX', ['rough sets', 'decision information'])\n", "('method used for task', 'a new method based on XXXXX is proposed to compute the XXXXX of a given graph', ['rough sets', 'minimum vertex cover'])\n", "('method used for task', 'presenting two algorithms for ivif XXXXX and XXXXX', ['incomplete information', 'importance weights'])\n", "('method used for task', 'XXXXX for the XXXXX parameters have been derived from the lyapunov–krasovskii functional candidate', ['adaptation laws', 'type-2 fuzzy logic system controller'])\n", "('method used for task', 'the congestion cost , rescheduled real power , XXXXX and XXXXX are less using hnm-fapso', ['power loss', 'computation time'])\n", "('NONE', 'the XXXXX , rescheduled real power , XXXXX and computation time are less using hnm-fapso', ['power loss', 'congestion cost'])\n", "('method used for task', 'the XXXXX , rescheduled real power , power loss and XXXXX are less using hnm-fapso', ['computation time', 'congestion cost'])\n", "('NONE', 'a multi-product multi-period XXXXX under XXXXX is developed', ['budget constraint', 'inventory control problem'])\n", "('NONE', 'a novel XXXXX selection algorithm that uses XXXXX for XXXXX selection', ['fuzzy logic', 'cluster head'])\n", "('NONE', 'a novel XXXXX algorithm that uses XXXXX for XXXXX', ['fuzzy logic', 'cluster head selection'])\n", "('NONE', 'a novel XXXXX algorithm that uses XXXXX for XXXXX', ['fuzzy logic', 'cluster head selection'])\n", "('method used for task', 'a novel XXXXX selection algorithm that uses fuzzy logic for XXXXX selection', ['cluster head', 'cluster head selection'])\n", "('method used for task', 'a novel XXXXX selection algorithm that uses fuzzy logic for XXXXX selection', ['cluster head', 'cluster head selection'])\n", "('method used for task', 'XXXXX is used to ensemble the XXXXX', ['weighted voting mechanism', 'base classifiers'])\n", "('method used for task', 'we propose a XXXXX model based on fuzzy XXXXX', ['logistic regression', 'technology credit scoring'])\n", "('method used for task', 'a novel XXXXX to forecast XXXXX prices was built', ['neuro-fuzzy controller', 'carbon emission'])\n", "('NONE', 'we have designed a XXXXX embedded device that would reduce the penalty billing of the consumers who fail to maintain the desired XXXXX', ['low cost', 'power factor'])\n", "('method used for task', 'switching of XXXXX banks though economical but often fail to give satisfactory compensation for varying load patterns . hence combination of XXXXX banks intelligently would give a better solution to this problem', ['shunt capacitor', 'shunt banks'])\n", "('method used for task', 'switching of XXXXX banks though economical but often fail to give satisfactory compensation for varying XXXXX . hence combination of XXXXX banks intelligently would give a better solution to this problem', ['shunt capacitor', 'load patterns'])\n", "('method used for task', 'switching of XXXXX banks though economical but often fail to give satisfactory compensation for varying load patterns . hence combination of XXXXX banks intelligently would give a better solution to this problem', ['shunt capacitor', 'shunt banks'])\n", "('method used for task', 'switching of XXXXX banks though economical but often fail to give satisfactory compensation for varying load patterns . hence combination of XXXXX banks intelligently would give a better solution to this problem', ['shunt banks', 'shunt capacitor'])\n", "('method used for task', 'switching of XXXXX banks though economical but often fail to give satisfactory compensation for varying XXXXX . hence combination of XXXXX banks intelligently would give a better solution to this problem', ['load patterns', 'shunt capacitor'])\n", "('method used for task', 'switching of XXXXX banks though economical but often fail to give satisfactory compensation for varying load patterns . hence combination of XXXXX banks intelligently would give a better solution to this problem', ['shunt capacitor', 'shunt banks'])\n", "('method used for task', 'we are presenting a modification of a XXXXX based on the bee behavior for optimizing XXXXX', ['bio-inspired algorithm', 'fuzzy controllers'])\n", "('NONE', 'XXXXX with XXXXX are used as benchmarks', ['vehicle routing problems', 'time windows'])\n", "('NONE', 'it starts the XXXXX with an XXXXX consisting of viruses and host cells . then , the environment will be clustered to number of regions called virus groups', ['initial population', 'optimization process'])\n", "('NONE', 'comparing the result of this algorithm with other well-known XXXXX demonstrates the superiority of the proposed algorithm in terms of the global solution convergence and the XXXXX', ['optimization algorithms', 'convergence speed'])\n", "('NONE', 'a two-level self-adaptive XXXXX is developed based on the two levels of decisions in the pcvrp , namely the selection of customers and the XXXXX of selected customers', ['vehicle scheduling', 'variable neighborhood search algorithm'])\n", "('method used for task', 'a graph extension method is adopted to obtain the XXXXX for pcvrp by transforming the proposed XXXXX of pcvrp into an equivalent traveling salesman problem model', ['lower bound', 'mixed integer programming model'])\n", "('NONE', 'XXXXX of XXXXX using legendre polynomial based functional link artificial neural network ( flann )', ['numerical solution', 'ordinary differential equations'])\n", "('NONE', 'the selection of XXXXX of ema is done through exhaustive XXXXX', ['control parameters', 'parametric study'])\n", "('method used for task', 'propose a new XXXXX to select best XXXXX', ['pareto front', 'decision making method'])\n", "('method used for task', 'in the present study , an XXXXX was used to solve a discrete mcr problem . during the solving process , the social force model was used to improve the XXXXX , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the XXXXX revealed that our algorithm was superior to the exact algorithm and greedy algorithm in terms of solution quality and running time', ['ant colony algorithm', 'simulation experiments'])\n", "('method used for task', 'in the present study , an XXXXX was used to solve a discrete mcr problem . during the solving process , the social force model was used to improve the XXXXX , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the simulation experiments revealed that our algorithm was superior to the XXXXX and greedy algorithm in terms of solution quality and running time', ['ant colony algorithm', 'exact algorithm'])\n", "('method used for task', 'in the present study , an XXXXX was used to solve a discrete mcr problem . during the solving process , the social force model was used to improve the XXXXX , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the simulation experiments revealed that our algorithm was superior to the exact algorithm and XXXXX in terms of solution quality and running time', ['ant colony algorithm', 'greedy algorithm'])\n", "('method used for task', 'in the present study , an XXXXX was used to solve a discrete mcr problem . during the solving process , the social force model was used to improve the XXXXX , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the simulation experiments revealed that our algorithm was superior to the exact algorithm and greedy algorithm in terms of solution quality and XXXXX', ['ant colony algorithm', 'running time'])\n", "('method used for task', 'in the present study , an XXXXX was used to solve a discrete mcr problem . during the solving process , the social XXXXX was used to improve the XXXXX , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the simulation experiments revealed that our algorithm was superior to the exact algorithm and greedy algorithm in terms of solution quality and running time', ['ant colony algorithm', 'force model'])\n", "('method used for task', 'in the present study , an XXXXX was used to solve a discrete mcr problem . during the solving process , the social force model was used to improve the XXXXX , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the simulation experiments revealed that our algorithm was superior to the exact algorithm and greedy algorithm in terms of XXXXX and running time', ['ant colony algorithm', 'solution quality'])\n", "('method used for task', 'in the present study , an XXXXX was used to solve a discrete mcr problem . during the solving process , the social force model was used to improve the XXXXX , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the XXXXX revealed that our algorithm was superior to the exact algorithm and greedy algorithm in terms of solution quality and running time', ['ant colony algorithm', 'simulation experiments'])\n", "('method used for task', 'in the present study , an XXXXX was used to solve a discrete mcr problem . during the solving process , the social force model was used to improve the XXXXX , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the simulation experiments revealed that our algorithm was superior to the XXXXX and greedy algorithm in terms of solution quality and running time', ['ant colony algorithm', 'exact algorithm'])\n", "('method used for task', 'in the present study , an XXXXX was used to solve a discrete mcr problem . during the solving process , the social force model was used to improve the XXXXX , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the simulation experiments revealed that our algorithm was superior to the exact algorithm and XXXXX in terms of solution quality and running time', ['ant colony algorithm', 'greedy algorithm'])\n", "('method used for task', 'in the present study , an XXXXX was used to solve a discrete mcr problem . during the solving process , the social force model was used to improve the XXXXX , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the simulation experiments revealed that our algorithm was superior to the exact algorithm and greedy algorithm in terms of solution quality and XXXXX', ['ant colony algorithm', 'running time'])\n", "('method used for task', 'in the present study , an XXXXX was used to solve a discrete mcr problem . during the solving process , the social XXXXX was used to improve the XXXXX , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the simulation experiments revealed that our algorithm was superior to the exact algorithm and greedy algorithm in terms of solution quality and running time', ['ant colony algorithm', 'force model'])\n", "('method used for task', 'in the present study , an XXXXX was used to solve a discrete mcr problem . during the solving process , the social force model was used to improve the XXXXX , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the simulation experiments revealed that our algorithm was superior to the exact algorithm and greedy algorithm in terms of XXXXX and running time', ['ant colony algorithm', 'solution quality'])\n", "('method used for task', 'in the present study , an ant colony algorithm was used to solve a discrete mcr problem . during the solving process , the social force model was used to improve the ant colony algorithm , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the XXXXX revealed that our algorithm was superior to the XXXXX and greedy algorithm in terms of solution quality and running time', ['simulation experiments', 'exact algorithm'])\n", "('method used for task', 'in the present study , an ant colony algorithm was used to solve a discrete mcr problem . during the solving process , the social force model was used to improve the ant colony algorithm , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the XXXXX revealed that our algorithm was superior to the exact algorithm and XXXXX in terms of solution quality and running time', ['simulation experiments', 'greedy algorithm'])\n", "('NONE', 'in the present study , an ant colony algorithm was used to solve a discrete mcr problem . during the solving process , the social force model was used to improve the ant colony algorithm , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the XXXXX revealed that our algorithm was superior to the exact algorithm and greedy algorithm in terms of solution quality and XXXXX', ['simulation experiments', 'running time'])\n", "('method used for task', 'in the present study , an ant colony algorithm was used to solve a discrete mcr problem . during the solving process , the social XXXXX was used to improve the ant colony algorithm , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the XXXXX revealed that our algorithm was superior to the exact algorithm and greedy algorithm in terms of solution quality and running time', ['simulation experiments', 'force model'])\n", "('NONE', 'in the present study , an ant colony algorithm was used to solve a discrete mcr problem . during the solving process , the social force model was used to improve the ant colony algorithm , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the XXXXX revealed that our algorithm was superior to the exact algorithm and greedy algorithm in terms of XXXXX and running time', ['simulation experiments', 'solution quality'])\n", "('method used for task', 'in the present study , an ant colony algorithm was used to solve a discrete mcr problem . during the solving process , the social force model was used to improve the ant colony algorithm , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the simulation experiments revealed that our algorithm was superior to the XXXXX and XXXXX in terms of solution quality and running time', ['exact algorithm', 'greedy algorithm'])\n", "('NONE', 'in the present study , an ant colony algorithm was used to solve a discrete mcr problem . during the solving process , the social force model was used to improve the ant colony algorithm , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the simulation experiments revealed that our algorithm was superior to the XXXXX and greedy algorithm in terms of solution quality and XXXXX', ['exact algorithm', 'running time'])\n", "('method used for task', 'in the present study , an ant colony algorithm was used to solve a discrete mcr problem . during the solving process , the social XXXXX was used to improve the ant colony algorithm , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the simulation experiments revealed that our algorithm was superior to the XXXXX and greedy algorithm in terms of solution quality and running time', ['exact algorithm', 'force model'])\n", "('NONE', 'in the present study , an ant colony algorithm was used to solve a discrete mcr problem . during the solving process , the social force model was used to improve the ant colony algorithm , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the simulation experiments revealed that our algorithm was superior to the XXXXX and greedy algorithm in terms of XXXXX and running time', ['exact algorithm', 'solution quality'])\n", "('NONE', 'in the present study , an ant colony algorithm was used to solve a discrete mcr problem . during the solving process , the social force model was used to improve the ant colony algorithm , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the simulation experiments revealed that our algorithm was superior to the exact algorithm and XXXXX in terms of solution quality and XXXXX', ['greedy algorithm', 'running time'])\n", "('method used for task', 'in the present study , an ant colony algorithm was used to solve a discrete mcr problem . during the solving process , the social XXXXX was used to improve the ant colony algorithm , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the simulation experiments revealed that our algorithm was superior to the exact algorithm and XXXXX in terms of solution quality and running time', ['greedy algorithm', 'force model'])\n", "('NONE', 'in the present study , an ant colony algorithm was used to solve a discrete mcr problem . during the solving process , the social force model was used to improve the ant colony algorithm , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the simulation experiments revealed that our algorithm was superior to the exact algorithm and XXXXX in terms of XXXXX and running time', ['greedy algorithm', 'solution quality'])\n", "('method used for task', 'in the present study , an ant colony algorithm was used to solve a discrete mcr problem . during the solving process , the social XXXXX was used to improve the ant colony algorithm , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the simulation experiments revealed that our algorithm was superior to the exact algorithm and greedy algorithm in terms of solution quality and XXXXX', ['running time', 'force model'])\n", "('method used for task', 'in the present study , an ant colony algorithm was used to solve a discrete mcr problem . during the solving process , the social force model was used to improve the ant colony algorithm , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the simulation experiments revealed that our algorithm was superior to the exact algorithm and greedy algorithm in terms of XXXXX and XXXXX', ['running time', 'solution quality'])\n", "('method used for task', 'in the present study , an ant colony algorithm was used to solve a discrete mcr problem . during the solving process , the social XXXXX was used to improve the ant colony algorithm , such that it would not easily fall into local extreme and became suitable for solving the mcr problem . the results of the simulation experiments revealed that our algorithm was superior to the exact algorithm and greedy algorithm in terms of XXXXX and running time', ['force model', 'solution quality'])\n", "('method used for task', 'scd is predicated using XXXXX and sudden XXXXX index ( scdi )', ['svm classifier', 'cardiac death'])\n", "('NONE', 'XXXXX are extracted from XXXXX', ['nonlinear features', 'hrv signals'])\n", "('method used for task', 'a novel de-noising method is proposed to remove the ecg interference from noisy emgdi signals . experimental results achieved on practical XXXXX show that the proposed approach is better than traditional methods include XXXXX ( wt ) , ica , digital filter and adaptive filter in ecg interference removing', ['clinical data', 'wavelet transform'])\n", "('method used for task', 'a novel de-noising method is proposed to remove the ecg interference from noisy emgdi signals . experimental results achieved on practical XXXXX show that the proposed approach is better than traditional methods include wavelet transform ( wt ) , ica , XXXXX and adaptive filter in ecg interference removing', ['clinical data', 'digital filter'])\n", "('NONE', 'a novel de-noising method is proposed to remove the ecg interference from noisy emgdi signals . experimental results achieved on practical clinical data show that the proposed approach is better than traditional methods include XXXXX ( wt ) , ica , XXXXX and adaptive filter in ecg interference removing', ['wavelet transform', 'digital filter'])\n", "('NONE', 'the original XXXXX of contaminated emgdi signal were firstly obtained with independent XXXXX ( ica ) . then the ecg components contained were removed by a specially designed wavelet domain filter . finally , the purified XXXXX were reconstructed back to the original signal space by ica to obtain the clean emgdi signals', ['independent components', 'component analysis'])\n", "('NONE', 'the original XXXXX of contaminated emgdi signal were firstly obtained with independent component analysis ( ica ) . then the ecg components contained were removed by a specially designed XXXXX filter . finally , the purified XXXXX were reconstructed back to the original signal space by ica to obtain the clean emgdi signals', ['independent components', 'wavelet domain'])\n", "('NONE', 'the original XXXXX of contaminated emgdi signal were firstly obtained with independent XXXXX ( ica ) . then the ecg components contained were removed by a specially designed wavelet domain filter . finally , the purified XXXXX were reconstructed back to the original signal space by ica to obtain the clean emgdi signals', ['independent components', 'component analysis'])\n", "('NONE', 'the original XXXXX of contaminated emgdi signal were firstly obtained with independent component analysis ( ica ) . then the ecg components contained were removed by a specially designed XXXXX filter . finally , the purified XXXXX were reconstructed back to the original signal space by ica to obtain the clean emgdi signals', ['independent components', 'wavelet domain'])\n", "('NONE', 'the original independent components of contaminated emgdi signal were firstly obtained with independent XXXXX ( ica ) . then the ecg components contained were removed by a specially designed XXXXX filter . finally , the purified independent components were reconstructed back to the original signal space by ica to obtain the clean emgdi signals', ['component analysis', 'wavelet domain'])\n", "('method used for task', 'the classification is performed using two different classifiers ; XXXXX and XXXXX', ['support vector machine', 'k-nearest neighbors'])\n", "('NONE', 'the actual plant input is known to the controller at each XXXXX , therefore , it can cope with relatively large network delays and XXXXX', ['time step', 'packet losses'])\n", "('NONE', 'shown improved XXXXX using battery as a XXXXX', ['prediction accuracy', 'case study'])\n", "('NONE', 'we propose a XXXXX for elm using XXXXX ( ga )', ['genetic algorithms', 'pruning method'])\n", "('NONE', 'this paper proposes a new XXXXX based on rules of XXXXX defined within the framework of cell-like p-systems to solve sudoku', ['particle swarm optimization', 'membrane algorithm'])\n", "('method used for task', 'also , sudoku of XXXXX is solved for which the XXXXX is about 87 %', ['higher order', 'success rate'])\n", "('NONE', 'obtain good XXXXX in terms of the solution quality and XXXXX', ['convergence speed', 'reconstruction performance'])\n", "('method used for task', 'obtain good reconstruction performance in terms of the XXXXX and XXXXX', ['convergence speed', 'solution quality'])\n", "('NONE', 'obtain good XXXXX in terms of the XXXXX and convergence speed', ['reconstruction performance', 'solution quality'])\n", "('method used for task', 'the novel XXXXX for the XXXXX', ['genetic algorithm', 'protein structure prediction problem'])\n", "('NONE', 'our XXXXX is developed by the adoption of the XXXXX', ['incremental algorithm', 'attribute reduction process'])\n", "('NONE', 'this paper presents an approach for XXXXX in digital mammograms using XXXXX', ['breast cancer diagnosis', 'shearlet transform'])\n", "('NONE', 'a new approach is represented for XXXXX using XXXXX , t-test statistic and dynamic thresholding', ['feature extraction', 'shearlet transform'])\n", "('NONE', 'we analyze the behavior of this XXXXX under the presence of XXXXX', ['neural network', 'class imbalance'])\n", "('method used for task', 'a heuristic ant colony optimization approach with adjustment strategy is proposed for the bearing surface packing problem so that the XXXXX is reduced from real space to integer space and the XXXXX is improved obviously', ['search space', 'computational efficiency'])\n", "('NONE', 'a XXXXX ( pso ) approach is designed to optimize the mass center and inertia angles of the satellite module in a way of rotation so that the XXXXX is improved significantly', ['particle swarm optimization', 'solution quality'])\n", "('method used for task', 'the solution quality of the proposed hybrid multi-mechanism optimization approach ( hmmoa ) is better than those of the dual-system variable-grain cooperative co-evolutionary XXXXX ( dvgccga ) and XXXXX through an organism combination of the quasi-principal component analysis , XXXXX and pso ( qpgp )', ['genetic algorithm', 'hybrid method'])\n", "('NONE', 'the XXXXX of the proposed hybrid multi-mechanism optimization approach ( hmmoa ) is better than those of the dual-system variable-grain cooperative co-evolutionary XXXXX ( dvgccga ) and hybrid method through an organism combination of the quasi-principal component analysis , XXXXX and pso ( qpgp )', ['genetic algorithm', 'solution quality'])\n", "('NONE', 'the solution quality of the proposed hybrid multi-mechanism XXXXX ( hmmoa ) is better than those of the dual-system variable-grain cooperative co-evolutionary XXXXX ( dvgccga ) and hybrid method through an organism combination of the quasi-principal component analysis , XXXXX and pso ( qpgp )', ['genetic algorithm', 'optimization approach'])\n", "('NONE', 'the solution quality of the proposed hybrid multi-mechanism optimization approach ( hmmoa ) is better than those of the dual-system variable-grain cooperative co-evolutionary XXXXX ( dvgccga ) and hybrid method through an organism combination of the quasi-principal XXXXX , XXXXX and pso ( qpgp )', ['genetic algorithm', 'component analysis'])\n", "('method used for task', 'the solution quality of the proposed hybrid multi-mechanism optimization approach ( hmmoa ) is better than those of the dual-system variable-grain cooperative co-evolutionary XXXXX ( dvgccga ) and XXXXX through an organism combination of the quasi-principal component analysis , XXXXX and pso ( qpgp )', ['hybrid method', 'genetic algorithm'])\n", "('NONE', 'the XXXXX of the proposed hybrid multi-mechanism optimization approach ( hmmoa ) is better than those of the dual-system variable-grain cooperative co-evolutionary genetic algorithm ( dvgccga ) and XXXXX through an organism combination of the quasi-principal component analysis , genetic algorithm and pso ( qpgp )', ['hybrid method', 'solution quality'])\n", "('method used for task', 'the solution quality of the proposed hybrid multi-mechanism XXXXX ( hmmoa ) is better than those of the dual-system variable-grain cooperative co-evolutionary genetic algorithm ( dvgccga ) and XXXXX through an organism combination of the quasi-principal component analysis , genetic algorithm and pso ( qpgp )', ['hybrid method', 'optimization approach'])\n", "('NONE', 'the solution quality of the proposed hybrid multi-mechanism optimization approach ( hmmoa ) is better than those of the dual-system variable-grain cooperative co-evolutionary genetic algorithm ( dvgccga ) and XXXXX through an organism combination of the quasi-principal XXXXX , genetic algorithm and pso ( qpgp )', ['hybrid method', 'component analysis'])\n", "('NONE', 'the XXXXX of the proposed hybrid multi-mechanism optimization approach ( hmmoa ) is better than those of the dual-system variable-grain cooperative co-evolutionary XXXXX ( dvgccga ) and hybrid method through an organism combination of the quasi-principal component analysis , XXXXX and pso ( qpgp )', ['genetic algorithm', 'solution quality'])\n", "('NONE', 'the solution quality of the proposed hybrid multi-mechanism XXXXX ( hmmoa ) is better than those of the dual-system variable-grain cooperative co-evolutionary XXXXX ( dvgccga ) and hybrid method through an organism combination of the quasi-principal component analysis , XXXXX and pso ( qpgp )', ['genetic algorithm', 'optimization approach'])\n", "('NONE', 'the solution quality of the proposed hybrid multi-mechanism optimization approach ( hmmoa ) is better than those of the dual-system variable-grain cooperative co-evolutionary XXXXX ( dvgccga ) and hybrid method through an organism combination of the quasi-principal XXXXX , XXXXX and pso ( qpgp )', ['genetic algorithm', 'component analysis'])\n", "('NONE', 'the XXXXX of the proposed hybrid multi-mechanism XXXXX ( hmmoa ) is better than those of the dual-system variable-grain cooperative co-evolutionary genetic algorithm ( dvgccga ) and hybrid method through an organism combination of the quasi-principal component analysis , genetic algorithm and pso ( qpgp )', ['solution quality', 'optimization approach'])\n", "('NONE', 'the XXXXX of the proposed hybrid multi-mechanism optimization approach ( hmmoa ) is better than those of the dual-system variable-grain cooperative co-evolutionary genetic algorithm ( dvgccga ) and hybrid method through an organism combination of the quasi-principal XXXXX , genetic algorithm and pso ( qpgp )', ['solution quality', 'component analysis'])\n", "('NONE', 'the solution quality of the proposed hybrid multi-mechanism XXXXX ( hmmoa ) is better than those of the dual-system variable-grain cooperative co-evolutionary genetic algorithm ( dvgccga ) and hybrid method through an organism combination of the quasi-principal XXXXX , genetic algorithm and pso ( qpgp )', ['optimization approach', 'component analysis'])\n", "('NONE', 'a new ipd model of XXXXX with adaptive XXXXX was built', ['multiple agents', 'risk attitudes'])\n", "('method used for task', 'XXXXX and XXXXX provide the best intervals', ['genetic programming', 'linear regression'])\n", "('method used for task', 'this paper studies the XXXXX and XXXXX with class setups in a single machine environment', ['order acceptance', 'scheduling problem'])\n", "('NONE', 'this XXXXX the XXXXX and scheduling problem with class setups in a single machine environment', ['order acceptance', 'paper studies'])\n", "('method used for task', 'this XXXXX the order acceptance and XXXXX with class setups in a single machine environment', ['scheduling problem', 'paper studies'])\n", "('NONE', 'a hybrid XXXXX in a XXXXX is proposed to solve the aircraft landing problem', ['particle swarm optimization algorithm', 'rolling horizon framework'])\n", "('method used for task', 'the main contributions of this paper are a strategy based on pso and XXXXX for optimizing a XXXXX that responds stronger to multi-scale and multi-orientation texture micro-patterns in a mammogram and enhances the classification rate', ['incremental clustering', 'gabor filter bank'])\n", "('method used for task', 'the main contributions of this paper are a strategy based on pso and XXXXX for optimizing a gabor filter bank that responds stronger to multi-scale and multi-orientation texture micro-patterns in a mammogram and enhances the XXXXX', ['incremental clustering', 'classification rate'])\n", "('method used for task', 'the main contributions of this paper are a strategy based on pso and incremental clustering for optimizing a XXXXX that responds stronger to multi-scale and multi-orientation texture micro-patterns in a mammogram and enhances the XXXXX', ['gabor filter bank', 'classification rate'])\n", "('method used for task', 'for the evaluation of the effect of the optimized XXXXX on the mass classification problems , it is applied on overlapping blocks of the rois to collect moment-based features ( i.e. , mean , XXXXX , skewness ) from the magnitudes of gabor responses', ['gabor filter bank', 'standard deviation'])\n", "('NONE', 'we focus on XXXXX ( mc ) method in XXXXX', ['monte carlo', 'fuzzy linear regression models'])\n", "('NONE', 'the study covers different XXXXX that have not previously calculated for monte carlo study in XXXXX', ['error measures', 'fuzzy linear regression models'])\n", "('method used for task', 'we obtain the most useful and the worst XXXXX to estimate fuzzy regression parameters without using any XXXXX or heavy fuzzy arithmetic operations', ['error measures', 'mathematical programming'])\n", "('NONE', 'the tsk-type XXXXX uses flann in the consequent part of the XXXXX for improved mapping', ['fuzzy inference system', 'fuzzy rules'])\n", "('NONE', 'the XXXXX , asymmetric shock by XXXXX of XXXXX are important for forecasting', ['leverage effect', 'egarch model'])\n", "('NONE', 'the XXXXX , asymmetric shock by XXXXX of XXXXX are important for forecasting', ['leverage effect', 'egarch model'])\n", "('NONE', 'a XXXXX based XXXXX is used for egarch-flann model parameters', ['differential evolution', 'learning strategy'])\n", "('method used for task', 'a XXXXX based learning strategy is used for egarch-flann XXXXX', ['differential evolution', 'model parameters'])\n", "('method used for task', 'a differential evolution based XXXXX is used for egarch-flann XXXXX', ['learning strategy', 'model parameters'])\n", "('NONE', 'the proposed method outperformed the conventional methods ( the multi-layer autoencoder , the denoising autoencoder , and kernel pca ) for the XXXXX on XXXXX', ['synthetic data', 'linear regression task'])\n", "('NONE', 'in the XXXXX on ion conductivity data of bulk materials and hydrogen storage materials , the good XXXXX was obtained in terms of the latent data uniformed by the proposed method', ['linear regression task', 'fitting performance'])\n", "('NONE', 'proposed methodology uses the current amplitudes signal in XXXXX as the inputs of the XXXXX for fault diagnosis', ['time domain', 'multi-agent system'])\n", "('NONE', 'proposed methodology uses the current amplitudes signal in XXXXX as the inputs of the multi-agent system for XXXXX', ['time domain', 'fault diagnosis'])\n", "('method used for task', 'proposed methodology uses the current amplitudes signal in time domain as the inputs of the XXXXX for XXXXX', ['multi-agent system', 'fault diagnosis'])\n", "('NONE', 'the XXXXX incorporates XXXXX with better results for each type of fault', ['multi-agent system', 'pattern recognition techniques'])\n", "('NONE', 'experimental results gathered from three-phase XXXXX operating with different XXXXX and fed under unbalance voltage are provided', ['induction motors', 'load conditions'])\n", "('method used for task', 'we propose a XXXXX based on the XXXXX and semivalue', ['betweenness centrality', 'shapley value'])\n", "('NONE', 'we provide an XXXXX of algorithms on real-life and XXXXX', ['empirical evaluation', 'random graphs'])\n", "('NONE', 'we present an architecture based on XXXXX that allows to build a portable classifier with XXXXX , and we reach state-of-the-art performance', ['feature selection', 'minimum cost'])\n", "('NONE', 'XXXXX exploits XXXXX and all available mined information', ['domain knowledge', 'process comparison'])\n", "('NONE', 'tests in XXXXX show that , with respect to previously published metrics our approach generates outputs closer to those of a XXXXX expert ; the framework can support experts in answering key XXXXX', ['stroke management', 'research questions'])\n", "('NONE', 'tests in XXXXX show that , with respect to previously published metrics our approach generates outputs closer to those of a XXXXX expert ; the framework can support experts in answering key XXXXX', ['stroke management', 'research questions'])\n", "('NONE', 'XXXXX of common entries and similar XXXXX in two different pathways', ['automatic identification', 'reaction paths'])\n", "('method used for task', 'the proposed methodology , visual diagnostic dss , employs ml algorithms and XXXXX for XXXXX in medical genetics', ['automated diagnosis', 'image processing techniques'])\n", "('method used for task', 'nicesim provides a XXXXX for simulation after XXXXX', ['graphical user interface', 'model building'])\n", "('method used for task', 'we focus on short-term XXXXX considering complete surgery flow and the XXXXX involved', ['multiple resources', 'surgery scheduling problem'])\n", "('NONE', 'we propose a modified XXXXX with a two-level ant XXXXX and specific aco mechanism', ['graph model', 'aco algorithm'])\n", "('NONE', 'we proposed improved XXXXX measures of simplified neutrosophic sets ( snss ) based on cosine function , including single valued neutrosophic XXXXX measures and interval neutrosophic XXXXX measures , to overcome some disadvantages of existing XXXXX measures of snss', ['cosine similarity', 'cosine similarity measures'])\n", "('NONE', 'we proposed improved XXXXX measures of simplified neutrosophic sets ( snss ) based on cosine function , including single valued neutrosophic XXXXX measures and interval neutrosophic XXXXX measures , to overcome some disadvantages of existing XXXXX measures of snss', ['cosine similarity', 'cosine similarity measures'])\n", "('NONE', 'we proposed improved XXXXX measures of simplified neutrosophic sets ( snss ) based on cosine function , including single valued neutrosophic XXXXX measures and interval neutrosophic XXXXX measures , to overcome some disadvantages of existing XXXXX measures of snss', ['cosine similarity', 'cosine similarity measures'])\n", "('NONE', 'we proposed improved XXXXX measures of simplified neutrosophic sets ( snss ) based on cosine function , including single valued neutrosophic XXXXX measures and interval neutrosophic XXXXX measures , to overcome some disadvantages of existing XXXXX measures of snss', ['cosine similarity', 'cosine similarity measures'])\n", "('NONE', 'we proposed improved XXXXX measures of simplified neutrosophic sets ( snss ) based on cosine function , including single valued neutrosophic XXXXX measures and interval neutrosophic XXXXX measures , to overcome some disadvantages of existing XXXXX measures of snss', ['cosine similarity', 'cosine similarity measures'])\n", "('NONE', 'we proposed improved XXXXX measures of simplified neutrosophic sets ( snss ) based on cosine function , including single valued neutrosophic XXXXX measures and interval neutrosophic XXXXX measures , to overcome some disadvantages of existing XXXXX measures of snss', ['cosine similarity', 'cosine similarity measures'])\n", "('NONE', 'we proposed improved XXXXX measures of simplified neutrosophic sets ( snss ) based on cosine function , including single valued neutrosophic XXXXX measures and interval neutrosophic XXXXX measures , to overcome some disadvantages of existing XXXXX measures of snss', ['cosine similarity', 'cosine similarity measures'])\n", "('NONE', 'we proposed improved XXXXX measures of simplified neutrosophic sets ( snss ) based on cosine function , including single valued neutrosophic XXXXX measures and interval neutrosophic XXXXX measures , to overcome some disadvantages of existing XXXXX measures of snss', ['cosine similarity', 'cosine similarity measures'])\n", "('NONE', 'we proposed improved XXXXX measures of simplified neutrosophic sets ( snss ) based on cosine function , including single valued neutrosophic XXXXX measures and interval neutrosophic XXXXX measures , to overcome some disadvantages of existing XXXXX measures of snss', ['cosine similarity', 'cosine similarity measures'])\n", "('NONE', 'we proposed improved XXXXX measures of simplified neutrosophic sets ( snss ) based on cosine function , including single valued neutrosophic XXXXX measures and interval neutrosophic XXXXX measures , to overcome some disadvantages of existing XXXXX measures of snss', ['cosine similarity', 'cosine similarity measures'])\n", "('NONE', 'we proposed improved XXXXX measures of simplified neutrosophic sets ( snss ) based on cosine function , including single valued neutrosophic XXXXX measures and interval neutrosophic XXXXX measures , to overcome some disadvantages of existing XXXXX measures of snss', ['cosine similarity', 'cosine similarity measures'])\n", "('NONE', 'we proposed improved XXXXX measures of simplified neutrosophic sets ( snss ) based on cosine function , including single valued neutrosophic XXXXX measures and interval neutrosophic XXXXX measures , to overcome some disadvantages of existing XXXXX measures of snss', ['cosine similarity', 'cosine similarity measures'])\n", "('method used for task', 'we presented a XXXXX method based on the improved cosine similarity measures to solve XXXXX problems with simplified neutrosophic information', ['medical diagnosis', 'medical diagnosis method'])\n", "('method used for task', 'we presented a XXXXX method based on the improved XXXXX to solve XXXXX problems with simplified neutrosophic information', ['medical diagnosis', 'cosine similarity measures'])\n", "('method used for task', 'we presented a XXXXX method based on the improved cosine similarity measures to solve XXXXX problems with simplified neutrosophic information', ['medical diagnosis', 'medical diagnosis problems'])\n", "('method used for task', 'we presented a XXXXX based on the improved XXXXX to solve medical diagnosis problems with simplified neutrosophic information', ['medical diagnosis method', 'cosine similarity measures'])\n", "('method used for task', 'we presented a XXXXX based on the improved cosine similarity measures to solve XXXXX with simplified neutrosophic information', ['medical diagnosis method', 'medical diagnosis problems'])\n", "('method used for task', 'we presented a medical diagnosis method based on the improved XXXXX to solve XXXXX with simplified neutrosophic information', ['cosine similarity measures', 'medical diagnosis problems'])\n", "('method used for task', 'two XXXXX were given to show the effectiveness and rationality of the XXXXX using the improved cosine similarity measures', ['medical diagnosis problems', 'diagnosis method'])\n", "('NONE', 'two XXXXX were given to show the effectiveness and rationality of the diagnosis method using the improved XXXXX', ['medical diagnosis problems', 'cosine similarity measures'])\n", "('NONE', 'two medical diagnosis problems were given to show the effectiveness and rationality of the XXXXX using the improved XXXXX', ['diagnosis method', 'cosine similarity measures'])\n", "('method used for task', 'XXXXX ontology is a tool to create XXXXX of hci assessment', ['knowledge bases', 'investigation model'])\n", "('NONE', 'a semi-online XXXXX in the XXXXX is studied', ['patient scheduling problem', 'pathology laboratory'])\n", "('NONE', 'the approach reduces XXXXX of patients and improves XXXXX', ['waiting time', 'operations efficiency'])\n", "('method used for task', 'XXXXX approaches are evaluated for assigning XXXXX to emrs', ['supervised learning', 'diagnosis codes'])\n", "('NONE', 'we propose an approach to XXXXX in XXXXX based on grey relational analysis and d-s theory of evidence', ['fuzzy soft set', 'decision making'])\n", "('method used for task', 'we develop a XXXXX to integrate XXXXX with the basic rs model', ['hybrid model', 'word similarity'])\n", "('NONE', 'we improved XXXXX recall using suffix patterns evolved by XXXXX', ['genetic programming', 'drug ner'])\n", "('method used for task', 'prior XXXXX is incorporated into the XXXXX', ['domain knowledge', 'selection procedure'])\n", "('method used for task', 'knowledge engeering for medical expert systems dominated the first decade of aime , while XXXXX and XXXXX prevailed thereafter', ['machine learning', 'data mining'])\n", "('NONE', 'promising directions for XXXXX are XXXXX , personalized medicine , evidence based medicine , business process modeling , process mining and nlp in social media', ['future research', 'big data'])\n", "('NONE', 'promising directions for XXXXX are big data , XXXXX , evidence based medicine , business process modeling , process mining and nlp in social media', ['future research', 'personalized medicine'])\n", "('NONE', 'promising directions for XXXXX are big data , personalized medicine , evidence based medicine , business process modeling , XXXXX and nlp in social media', ['future research', 'process mining'])\n", "('NONE', 'promising directions for XXXXX are big data , personalized medicine , evidence based medicine , business process modeling , process mining and nlp in XXXXX', ['future research', 'social media'])\n", "('NONE', 'promising directions for future research are XXXXX , XXXXX , evidence based medicine , business process modeling , process mining and nlp in social media', ['big data', 'personalized medicine'])\n", "('NONE', 'promising directions for future research are XXXXX , personalized medicine , evidence based medicine , business process modeling , XXXXX and nlp in social media', ['big data', 'process mining'])\n", "('NONE', 'promising directions for future research are XXXXX , personalized medicine , evidence based medicine , business process modeling , process mining and nlp in XXXXX', ['big data', 'social media'])\n", "('NONE', 'promising directions for future research are big data , XXXXX , evidence based medicine , business process modeling , XXXXX and nlp in social media', ['personalized medicine', 'process mining'])\n", "('NONE', 'promising directions for future research are big data , XXXXX , evidence based medicine , business process modeling , process mining and nlp in XXXXX', ['personalized medicine', 'social media'])\n", "('NONE', 'promising directions for future research are big data , personalized medicine , evidence based medicine , business process modeling , XXXXX and nlp in XXXXX', ['process mining', 'social media'])\n", "('method used for task', 'XXXXX for simultaneous XXXXX and classifier weighting', ['evolutionary algorithm', 'feature selection'])\n", "('NONE', 'we encountered a clinical demand to process XXXXX within XXXXX medical logic modules', ['complex data', 'arden syntax'])\n", "('NONE', 'we encountered a clinical demand to process XXXXX within arden syntax medical XXXXX', ['complex data', 'logic modules'])\n", "('NONE', 'we encountered a clinical demand to process complex data within XXXXX medical XXXXX', ['arden syntax', 'logic modules'])\n", "('NONE', 'in 2004 we developed a pediatric computer based XXXXX , chica , using XXXXX medical logic modules ( mlm )', ['clinical decision support system', 'arden syntax'])\n", "('NONE', 'in 2004 we developed a pediatric computer based XXXXX , chica , using arden syntax medical XXXXX ( mlm )', ['clinical decision support system', 'logic modules'])\n", "('NONE', 'in 2004 we developed a pediatric computer based clinical decision support system , chica , using XXXXX medical XXXXX ( mlm )', ['arden syntax', 'logic modules'])\n", "('NONE', 'such a system can be used to combine data generated by XXXXX apps with XXXXX', ['clinical data', 'home monitoring'])\n", "('NONE', 'we examine its importance by an XXXXX of four XXXXX', ['empirical study', 'feature selection methods'])\n", "('method used for task', 'we use a XXXXX to improve XXXXX based on timing information', ['probabilistic model', 'classification accuracy'])\n", "('method used for task', 'the paper studies several XXXXX ( nlp ) techniques to extract predictors from uncoded data in XXXXX ( emrs )', ['natural language processing', 'electronic medical records'])\n", "('NONE', 'the XXXXX several XXXXX ( nlp ) techniques to extract predictors from uncoded data in electronic medical records ( emrs )', ['natural language processing', 'paper studies'])\n", "('method used for task', 'the XXXXX several natural language processing ( nlp ) techniques to extract predictors from uncoded data in XXXXX ( emrs )', ['electronic medical records', 'paper studies'])\n", "('NONE', 'the results show that some of the XXXXX studied can complement the coded emr data , and hence , result in improved XXXXX', ['predictive models', 'nlp techniques'])\n", "('method used for task', 'the objective is to provide real-world XXXXX and XXXXX', ['decision support', 'risk management'])\n", "('method used for task', 'both applications provide improvements in XXXXX and XXXXX', ['predictive accuracy', 'decision support'])\n", "('NONE', 'the developed XXXXX generates the parameters of a XXXXX by capturing the subjective and intuitive knowledge of medical physicists', ['cbr system', 'treatment plan'])\n", "('NONE', 'in this research we focus on the adaptation stage of the XXXXX in which the solution ( XXXXX ) of the retrieved case is adapted to meet the needs of the new case ( patient ) by considering differences between the retrieved and new cases', ['cbr system', 'treatment plan'])\n", "('NONE', 'the XXXXX was deployed alongside the XXXXX to evaluate its ability to discover rules that complement the existing rule set', ['learning module', 'baseline system'])\n", "('NONE', 'the XXXXX extracted clinically relevant rules , some of which extend the XXXXX of the baseline system', ['knowledge base', 'learning module'])\n", "('NONE', 'the learning module extracted clinically relevant rules , some of which extend the XXXXX of the XXXXX', ['knowledge base', 'baseline system'])\n", "('NONE', 'the XXXXX extracted clinically relevant rules , some of which extend the knowledge base of the XXXXX', ['learning module', 'baseline system'])\n", "('NONE', 'used XXXXX to determine multiple level of context in free living and contextualize XXXXX', ['machine learning methods', 'heart rate data'])\n", "('NONE', 'we develop a XXXXX with the XXXXX', ['particle filter', 'dirichlet distribution'])\n", "('NONE', 'XXXXX can be optimized by carefully XXXXX the workload', ['parallel performance', 'load balancing'])\n", "('method used for task', 'XXXXX is used for the XXXXX', ['comparative study', 'real time data'])\n", "('NONE', 'it integrates vo XXXXX as first-class XXXXX in taverna scientific workflows', ['web services', 'building blocks'])\n", "('NONE', 'the high XXXXX can be got based on XXXXX', ['clinical data', 'prediction performance'])\n", "('method used for task', 'a touching nuclei XXXXX based on XXXXX and concave vertex graph is proposed to quantify the total number of cancer nuclei in each class', ['watershed algorithm', 'separation method'])\n", "('NONE', 'XXXXX learn from XXXXX and multibody model solutions', ['neural networks', 'finite element'])\n", "('NONE', 'developed a novel framework to apply sparse and overcomplete representations for the removal of XXXXX in real XXXXX', ['speckle noise', 'ultrasound images'])\n", "('NONE', 'we proposed a combinational feature extraction approach based on some XXXXX extracted from electro cardio graph ( ecg ) XXXXX ( rps ) and usually used frequency domain features for detection of sleep apnoea', ['nonlinear features', 'reconstructed phase space'])\n", "('method used for task', 'here 6 XXXXX extracted from ecg rps are combined with 3 frequency based features to reconstruct final XXXXX', ['nonlinear features', 'feature set'])\n", "('NONE', 'the XXXXX consist of detrended fluctuation analysis ( dfa ) , correlation dimensions ( cd ) , 3 large XXXXX ( lles ) and spectral entropy ( se )', ['nonlinear features', 'lyapunov exponents'])\n", "('NONE', 'the XXXXX consist of detrended XXXXX ( dfa ) , correlation dimensions ( cd ) , 3 large lyapunov exponents ( lles ) and spectral entropy ( se )', ['nonlinear features', 'fluctuation analysis'])\n", "('NONE', 'the XXXXX consist of detrended fluctuation analysis ( dfa ) , XXXXX ( cd ) , 3 large lyapunov exponents ( lles ) and spectral entropy ( se )', ['nonlinear features', 'correlation dimensions'])\n", "('NONE', 'the nonlinear features consist of detrended XXXXX ( dfa ) , correlation dimensions ( cd ) , 3 large XXXXX ( lles ) and spectral entropy ( se )', ['lyapunov exponents', 'fluctuation analysis'])\n", "('NONE', 'the nonlinear features consist of detrended fluctuation analysis ( dfa ) , XXXXX ( cd ) , 3 large XXXXX ( lles ) and spectral entropy ( se )', ['lyapunov exponents', 'correlation dimensions'])\n", "('NONE', 'the nonlinear features consist of detrended XXXXX ( dfa ) , XXXXX ( cd ) , 3 large lyapunov exponents ( lles ) and spectral entropy ( se )', ['fluctuation analysis', 'correlation dimensions'])\n", "('NONE', 'a XXXXX ( fmm ) based XXXXX is proposed to interpolate the missing voxels', ['fast marching method', 'reconstruction algorithm'])\n", "('method used for task', 'the fmm XXXXX could preserve relatively sharper edges and more XXXXX in the reconstructed image slices', ['reconstruction algorithm', 'texture patterns'])\n", "('method used for task', 'the fmm XXXXX runs efficiently and requires less XXXXX', ['reconstruction algorithm', 'computation time'])\n", "('NONE', 'we classify different sleep stages with 41 XXXXX through XXXXX', ['random forest', 'hrv features'])\n", "('method used for task', 'we propose to extract texture feature descriptors based on XXXXX to characterize breast tumors in XXXXX', ['shearlet transform', 'ultrasound images'])\n", "('method used for task', 'novel XXXXX based on XXXXX and nonlinear transformation', ['feature extraction', 'tensor decomposition'])\n", "('NONE', 'the method can escape from XXXXX by using a fairly large XXXXX during freely deforming', ['local minima', 'detection range'])\n", "('method used for task', 'investigated XXXXX and model estimation methods for mr XXXXX', ['statistical models', 'image segmentation'])\n", "('method used for task', 'investigated XXXXX and XXXXX methods for mr image segmentation', ['statistical models', 'model estimation'])\n", "('method used for task', 'investigated statistical models and XXXXX methods for mr XXXXX', ['image segmentation', 'model estimation'])\n", "('NONE', 'XXXXX of eeg signal were extracted by XXXXX ( csp )', ['feature vectors', 'common spatial patterns'])\n", "('NONE', 'XXXXX were classified by XXXXX ( svm )', ['feature vectors', 'support vector machine'])\n", "('method used for task', 'we propose a multi-step directional ggvf XXXXX for XXXXX', ['snake model', 'target tumor segmentation'])\n", "('method used for task', 'for correlated rhythms , an XXXXX exists between bandwidth and XXXXX', ['spatial resolution', 'inverse relationship'])\n", "('method used for task', 'help the researchers to develop a new XXXXX for XXXXX', ['mr images', 'denoising method'])\n", "('method used for task', 'helps to choose the best XXXXX for further XXXXX', ['denoising method', 'image processing methods'])\n", "('NONE', 'XXXXX are XXXXX', ['respiratory sounds', 'chaotic signals'])\n", "('method used for task', 'a new XXXXX method is proposed for XXXXX in XXXXX', ['noise reduction', 'respiratory sounds'])\n", "('method used for task', 'a new XXXXX method is proposed for XXXXX in respiratory sounds', ['noise reduction', 'noise reduction method'])\n", "('method used for task', 'a new XXXXX is proposed for noise reduction in XXXXX', ['respiratory sounds', 'noise reduction method'])\n", "('NONE', 'the proposed method is based on new nonlinear XXXXX as controller that their parameter tunes by XXXXX', ['convex function', 'genetic algorithm'])\n", "('method used for task', 'a framework on wavelet-based XXXXX and extreme XXXXX is proposed for the seizure detection', ['nonlinear features', 'machine learning'])\n", "('method used for task', 'a framework on wavelet-based XXXXX and extreme machine learning is proposed for the XXXXX', ['nonlinear features', 'seizure detection'])\n", "('method used for task', 'a framework on wavelet-based nonlinear features and extreme XXXXX is proposed for the XXXXX', ['machine learning', 'seizure detection'])\n", "('method used for task', 'applications in the natural XXXXX and pathological XXXXX', ['speech synthesis', 'speech recognition'])\n", "('NONE', 'the paper presents a new method for XXXXX in ultrasound XXXXX', ['speckle reduction', 'medical images'])\n", "('method used for task', 'the proposed method combines the XXXXX and ripplet thresholding to remove the speckles in XXXXX', ['bilateral filter', 'ultrasound images'])\n", "('method used for task', 'this improved XXXXX is used as an input to evolve the initial XXXXX', ['edge information', 'level set function'])\n", "('NONE', 'we extend the method to afford XXXXX of new models in XXXXX', ['unsupervised learning', 'real time'])\n", "('NONE', \"a XXXXX estimated the time varying XXXXX representing the windkessel model 's output signal\", ['kalman filter', 'fourier series'])\n", "('method used for task', 'XXXXX reconstruct method is used to reconstruct the XXXXX of XXXXX', ['phase space', 'eeg signals'])\n", "('method used for task', 'XXXXX reconstruct method is used to reconstruct the XXXXX of XXXXX', ['phase space', 'eeg signals'])\n", "('NONE', 'the features are extracted by amplitude–frequency XXXXX in the XXXXX', ['phase space', 'analysis method'])\n", "('method used for task', 'using an fpga platform , we showed that the cs-based XXXXX , compared to a low-power wavelet-based XXXXX , can largely reduce XXXXX and save other computing resources', ['power consumption', 'compression method'])\n", "('method used for task', 'using an fpga platform , we showed that the cs-based XXXXX , compared to a low-power wavelet-based XXXXX , can largely reduce XXXXX and save other computing resources', ['power consumption', 'compression method'])\n", "('method used for task', 'design of an energy optimal trajectory planner , based on XXXXX to decide the optimal path for the XXXXX to move towards the target', ['differential evolution', 'robot arm'])\n", "('NONE', 'the XXXXX of the simulated XXXXX reaching the target is 85 %', ['robot arm', 'success rate'])\n", "('method used for task', 'we proposed a novel XXXXX rejection approach to suppress the XXXXX signals induced by the slowly moving vessel walls and tissues in composite XXXXX', ['doppler ultrasound signals', 'quadrature clutter'])\n", "('method used for task', 'we proposed a novel XXXXX rejection approach to suppress the XXXXX signals induced by the slowly moving vessel walls and tissues in composite XXXXX', ['doppler ultrasound signals', 'quadrature clutter'])\n", "('method used for task', 'the proposed XXXXX approach is based on multivariate empirical XXXXX ( memd )', ['clutter rejection', 'mode decomposition'])\n", "('NONE', 'the XXXXX between the quadrature demodulated XXXXX is keeping well during the decomposition by memd', ['phase information', 'doppler ultrasound signals'])\n", "('NONE', 'we quantify contrast agent XXXXX in plaque using XXXXX', ['spatial distribution', 'texture features'])\n", "('NONE', 'the XXXXX of histogram are extracted for XXXXX', ['statistical features', 'svm classification'])\n", "('NONE', 'a new angiogram background suppression method based on 3d XXXXX is proposed in the context of sparsity regularized XXXXX of coronary artery', ['vessel segmentation', 'iterative reconstruction'])\n", "('NONE', 'a new angiogram background suppression method based on 3d XXXXX is proposed in the context of sparsity regularized iterative reconstruction of XXXXX', ['vessel segmentation', 'coronary artery'])\n", "('NONE', 'a new angiogram background suppression method based on 3d vessel segmentation is proposed in the context of sparsity regularized XXXXX of XXXXX', ['iterative reconstruction', 'coronary artery'])\n", "('NONE', 'several experiments are performed to quantitatively evaluate the proposed method and experimental results show that it can effectively improve the XXXXX of XXXXX', ['coronary artery', 'reconstruction quality'])\n", "('NONE', 'XXXXX of XXXXX diabetic göttingen minipigs', ['model development', 'glucose metabolism'])\n", "('method used for task', 'development of XXXXX for future XXXXX', ['mathematical model', 'control design'])\n", "('NONE', 'XXXXX , especially the dynamic change of arm position , have adverse effect on surface electromyography pattern recognition of wrist or hand motions , which results in very high XXXXX', ['arm movements', 'classification errors'])\n", "('method used for task', 'we adopted three metrics to quantify the changes of characteristics in the XXXXX due to XXXXX', ['feature space', 'arm movements'])\n", "('method used for task', 'a robust training scheme was used to eliminate this effect and a novel classifier , conditional XXXXX , was proposed to improve the XXXXX under this training scheme', ['gaussian mixture model', 'classification performance'])\n", "('method used for task', 'a robust XXXXX was used to eliminate this effect and a novel classifier , conditional XXXXX , was proposed to improve the classification performance under this XXXXX', ['gaussian mixture model', 'training scheme'])\n", "('method used for task', 'a robust XXXXX was used to eliminate this effect and a novel classifier , conditional XXXXX , was proposed to improve the classification performance under this XXXXX', ['gaussian mixture model', 'training scheme'])\n", "('method used for task', 'a robust XXXXX was used to eliminate this effect and a novel classifier , conditional gaussian mixture model , was proposed to improve the XXXXX under this XXXXX', ['classification performance', 'training scheme'])\n", "('method used for task', 'a robust XXXXX was used to eliminate this effect and a novel classifier , conditional gaussian mixture model , was proposed to improve the XXXXX under this XXXXX', ['classification performance', 'training scheme'])\n", "('method used for task', 'boundaries of XXXXX ( s1 , s2 , s3 , s4 ) and murmurs are automatically determined using instantaneous XXXXX of the envelope', ['heart sounds', 'phase waveform'])\n", "('NONE', 'XXXXX from a contour integration task are analyzed with empirical XXXXX', ['fmri data', 'mode decomposition'])\n", "('NONE', 'introduction of a switching control method and XXXXX by means of two XXXXX of the porcine glucose metabolism', ['controller design', 'mathematical models'])\n", "('NONE', 'introduction of a switching control method and XXXXX by means of two mathematical models of the porcine XXXXX', ['controller design', 'glucose metabolism'])\n", "('method used for task', 'introduction of a XXXXX and XXXXX by means of two mathematical models of the porcine glucose metabolism', ['controller design', 'switching control method'])\n", "('NONE', 'introduction of a switching control method and controller design by means of two XXXXX of the porcine XXXXX', ['mathematical models', 'glucose metabolism'])\n", "('NONE', 'introduction of a XXXXX and controller design by means of two XXXXX of the porcine glucose metabolism', ['mathematical models', 'switching control method'])\n", "('NONE', 'introduction of a XXXXX and controller design by means of two mathematical models of the porcine XXXXX', ['glucose metabolism', 'switching control method'])\n", "('NONE', 'XXXXX of XXXXX by animal trials taking into account control performance and technical limitations', ['experimental evaluation', 'control method'])\n", "('NONE', 'XXXXX of control method by animal trials taking into account XXXXX and technical limitations', ['experimental evaluation', 'control performance'])\n", "('NONE', 'experimental evaluation of XXXXX by animal trials taking into account XXXXX and technical limitations', ['control method', 'control performance'])\n", "('method used for task', 'estimated parameters of the ar-garch model formed the new XXXXX for the XXXXX', ['feature set', 'pattern classification'])\n", "('NONE', 'the possibility to reconstruct a XXXXX from XXXXX in order to characterize it from a dynamical point of view is performed', ['mathematical model', 'experimental data'])\n", "('method used for task', 'the XXXXX , the fractal and XXXXX could be used for classification feature', ['time lag', 'correlation dimensions'])\n", "('NONE', 'the significance of the measurements is assessment based on the XXXXX , the total XXXXX ( the level of noise plus non-linearities ) , the experimental errors and the effect of using different sensors', ['standard deviation', 'noise level'])\n", "('NONE', 'the paper presents an algorithm for XXXXX in XXXXX', ['speckle reduction', 'ultrasound images'])\n", "('NONE', 'the proposed method is based on non-linear adaptive XXXXX in nonsubsampled XXXXX', ['diffusion model', 'shearlet domain'])\n", "('method used for task', 'the proposed method provides better XXXXX and edge preservation performance for ultrasound XXXXX as compared to others', ['noise suppression', 'medical images'])\n", "('NONE', 'we perform an XXXXX of XXXXX', ['automatic analysis', 'speech signals'])\n", "('NONE', 'XXXXX exhibit longer pauses between each XXXXX', ['pd patients', 'sentence repetition'])\n", "('method used for task', 'redundant information reduced significantly which saves XXXXX and improves XXXXX', ['computational time', 'image quality'])\n", "('NONE', 'we use a set of three XXXXX as a dynamical model of XXXXX', ['differential equations', 'ecg signals'])\n", "('method used for task', 'two classifiers are constructed to classify XXXXX according to the XXXXX', ['ecg signals', 'parameter changes'])\n", "('method used for task', 'the method uses XXXXX and XXXXX for the extraction of oriented patterns and linear structures representing breast tissues', ['gabor filters', 'radon transform'])\n", "('method used for task', 'XXXXX for the XXXXX based on sparse representation', ['compression algorithm', 'ecg signals'])\n", "('method used for task', 'XXXXX for the ecg signals based on XXXXX', ['compression algorithm', 'sparse representation'])\n", "('method used for task', 'compression algorithm for the XXXXX based on XXXXX', ['ecg signals', 'sparse representation'])\n", "('NONE', 'XXXXX are better in terms of power spectra and XXXXX', ['stochastic models', 'information entropy'])\n", "('NONE', 'XXXXX are better in terms of XXXXX and information entropy', ['stochastic models', 'power spectra'])\n", "('method used for task', 'stochastic models are better in terms of XXXXX and XXXXX', ['information entropy', 'power spectra'])\n", "('method used for task', 'bhattacharyya space algorithm , t-test , wilcoxon test , receiver operating curve ( roc ) , and XXXXX are used for XXXXX', ['feature ranking', 'entropy method'])\n", "('method used for task', 'the comparation is given between the XXXXX strategy and constant XXXXX , which shows the XXXXX strategy is better than constant XXXXX', ['optimal control', 'control strategy'])\n", "('method used for task', 'the comparation is given between the XXXXX strategy and constant control strategy , which shows the XXXXX strategy is better than constant control strategy', ['optimal control', 'optimal control strategy'])\n", "('method used for task', 'the comparation is given between the XXXXX strategy and constant control strategy , which shows the XXXXX strategy is better than constant control strategy', ['optimal control', 'optimal control strategy'])\n", "('method used for task', 'the comparation is given between the optimal XXXXX and constant XXXXX , which shows the optimal XXXXX is better than constant XXXXX', ['control strategy', 'optimal control strategy'])\n", "('method used for task', 'the comparation is given between the optimal XXXXX and constant XXXXX , which shows the optimal XXXXX is better than constant XXXXX', ['control strategy', 'optimal control strategy'])\n", "('NONE', 'increasing XXXXX parameter will decrease the XXXXX', ['model performance', 'constraint window'])\n", "('NONE', 'stochastic model predictive ( stomp ) is a glycaemic XXXXX that combines the probabilistic , stochastic XXXXX of previous methods ( star ) with model predictive control , for ease of tuning', ['control protocol', 'forecasting methods'])\n", "('method used for task', 'this paper presents a hybrid XXXXX for ultrasound XXXXX', ['medical images', 'segmentation method'])\n", "('NONE', 'the proposed algorithm can distinguish among six classes of XXXXX using two temporal eeg sensors with XXXXX and low-latency using a single-trial to make a decision', ['eye movements', 'high accuracy'])\n", "('NONE', 'XXXXX ( acms ) detecting XXXXX ( od ) boundaries are investigated', ['active contour models', 'optic disc'])\n", "('method used for task', 'images are subject to adaptive XXXXX and XXXXX', ['histogram equalization', 'morphological operations'])\n", "('NONE', 'we achieve better performance on simulated and XXXXX medical XXXXX', ['in vivo', 'ultrasound images'])\n", "('method used for task', 'we develop ground reaction force ( grf ) based XXXXX for XXXXX', ['human gait', 'asymmetry measure'])\n", "('method used for task', 'the proposed method is based on the XXXXX measure and the pcnn model for fusing the image coefficients in nonsubsampled XXXXX', ['activity level', 'shearlet domain'])\n", "('method used for task', 'cleaned real XXXXX are used for pattern recognition-based XXXXX', ['gesture recognition', 'emg signals'])\n", "('method used for task', 'XXXXX is improved for real emg with added XXXXX', ['classification accuracy', 'white noise'])\n", "('NONE', 'a new algorithm named ibfsvm is proposed , which is based on fuzzy XXXXX and can be used in XXXXX', ['support vector', 'imbalanced classification'])\n", "('method used for task', 'the multi-ganglion ann based XXXXX ( annfl ) method is an unsupervised XXXXX', ['feature learning', 'feature extraction method'])\n", "('method used for task', 'a novel method based on XXXXX is proposed for XXXXX from eeg data , simultaneously recorded with fmri', ['dictionary learning', 'bcg removal'])\n", "('NONE', 'a new XXXXX , having the generic XXXXX as its core , is proposed', ['cost function', 'dictionary learning'])\n", "('method used for task', 'the hms based entropy and XXXXX are extracted for XXXXX', ['svm classification', 'energy features'])\n", "('method used for task', 'an optimal XXXXX is chosen by formal XXXXX', ['feature vector', 'statistical testing'])\n", "('method used for task', 'we developed a new XXXXX to detect XXXXX from ecg', ['statistical model', 'atrial fibrillation'])\n", "('NONE', 'an automatic detection algorithm for detection of pathologic voices is suggested . residual signals extracted by applying XXXXX on wavelet sub-bands provide a highly reliable vocal disorders detection scheme . by employing XXXXX and support vector machine , a great classification is achieved', ['linear prediction', 'linear discriminant analysis'])\n", "('method used for task', 'an automatic detection algorithm for detection of pathologic voices is suggested . residual signals extracted by applying XXXXX on wavelet sub-bands provide a highly reliable vocal disorders detection scheme . by employing linear discriminant analysis and XXXXX , a great classification is achieved', ['linear prediction', 'support vector machine'])\n", "('method used for task', 'an automatic detection algorithm for detection of pathologic voices is suggested . residual signals extracted by applying linear prediction on wavelet sub-bands provide a highly reliable vocal disorders detection scheme . by employing XXXXX and XXXXX , a great classification is achieved', ['linear discriminant analysis', 'support vector machine'])\n", "('NONE', 'motivation : synchronize XXXXX from several XXXXX is a difficult task', ['biomedical signals', 'acquisition systems'])\n", "('method used for task', 'methodology : use of XXXXX and correlation methods to synchronize XXXXX', ['white noise', 'biomedical signals'])\n", "('method used for task', 'XXXXX largely compresses neural signals and has a small XXXXX', ['reconstruction error', 'mdc matrix'])\n", "('NONE', 'XXXXX can be used over a large range of the XXXXX for compression', ['mdc matrix', 'sampling rate'])\n", "('method used for task', 'an improved XXXXX for multimodal XXXXX', ['retinal images', 'registration framework'])\n", "('method used for task', 'we propose two adaptive compression paradigms for the XXXXX ( dwt ) and XXXXX ( cs ) compression techniques', ['discrete wavelet transform', 'compressive sensing'])\n", "('method used for task', 'we propose two adaptive compression paradigms for the XXXXX ( dwt ) and compressive sensing ( cs ) XXXXX', ['discrete wavelet transform', 'compression techniques'])\n", "('NONE', 'we propose two adaptive compression paradigms for the discrete wavelet transform ( dwt ) and XXXXX ( cs ) XXXXX', ['compressive sensing', 'compression techniques'])\n", "('NONE', 'the paper presents a XXXXX of newborn cry signals based on XXXXX', ['hidden markov models', 'segmentation system'])\n", "('method used for task', 'the XXXXX and respiratory rate are estimated form the XXXXX envelope', ['heart rate', 'shannon energy'])\n", "('method used for task', 'a XXXXX for eeg XXXXX is proposed which combines permutation and lempel-ziv complexity ( plzc )', ['signal analysis', 'complexity measure'])\n", "('method used for task', 'combine XXXXX and wilcoxon statistics for eeg signal XXXXX', ['wavelet transform', 'feature extraction'])\n", "('method used for task', 'pulsations of XXXXX and XXXXX are correlated for 0.01–0.1hz', ['blood flow', 'skin temperature'])\n", "('NONE', 'finding best samples for lmmse estimation in XXXXX will improve the XXXXX', ['dct domain', 'denoising performance'])\n", "('method used for task', 'its performances were tested using both coupled XXXXX and clinical cardiovascular coupled signals , and were also compared with two existing cross entropy measures : cross XXXXX ( c-sampen ) and cross fuzzy entropy ( c-fuzzyen )', ['simulation models', 'sample entropy'])\n", "('method used for task', 'its performances were tested using both coupled XXXXX and clinical cardiovascular coupled signals , and were also compared with two existing cross entropy measures : cross sample entropy ( c-sampen ) and cross XXXXX ( c-fuzzyen )', ['simulation models', 'fuzzy entropy'])\n", "('method used for task', 'its performances were tested using both coupled simulation models and clinical cardiovascular coupled signals , and were also compared with two existing cross entropy measures : cross XXXXX ( c-sampen ) and cross XXXXX ( c-fuzzyen )', ['sample entropy', 'fuzzy entropy'])\n", "('NONE', 'XXXXX ( hr ) monitoring using wrist-type XXXXX during physical exercise was examined', ['heart rate', 'ppg signals'])\n", "('NONE', 'evaluation of XXXXX of XXXXX based classification ( src ) method', ['noise robustness', 'sparse representation'])\n", "('NONE', 'generation of new noisy XXXXX by adding two XXXXX into the eeg XXXXX', ['test signals', 'noise sources'])\n", "('NONE', 'generation of new noisy XXXXX by adding two XXXXX into the eeg XXXXX', ['noise sources', 'test signals'])\n", "('NONE', 'demonstration of the improved XXXXX of the src for noisy XXXXX and online dataset', ['classification accuracy', 'test signals'])\n", "('NONE', 'XXXXX increased with XXXXX', ['image quality', 'compressive sensing method'])\n", "('NONE', 'we use 3d-dwt to capture the 3d XXXXX of brain , use als-pca for XXXXX of dataset containing missing attributes', ['texture feature', 'feature reduction'])\n", "('NONE', 'an unsupervised , XXXXX , XXXXX for brain tumor segmentation is proposed', ['low cost', 'hybrid approach'])\n", "('method used for task', 'XXXXX is evaluated by standard viability assay while XXXXX required the development of a special algorithm', ['cell proliferation', 'cell morphology'])\n", "('NONE', 'our data show that ultrasounds induce cell differentiation affecting XXXXX as well as the ability of neurons and microglia to form XXXXX', ['complex networks', 'cell morphology'])\n", "('method used for task', 'we propose a multiscale svd based XXXXX for multilead XXXXX', ['compression algorithm', 'ecg data'])\n", "('method used for task', 'XXXXX and XXXXX are employed to improve performance of the system', ['genetic algorithm', 'linear discriminant analysis'])\n", "('NONE', 'XXXXX permit identifying characteristic points of the XXXXX', ['curvelet transform', 'ecg signal'])\n", "('NONE', 'detection of pvc in XXXXX using the fractional XXXXX outperforms detection using XXXXX', ['ecg signals', 'linear prediction'])\n", "('NONE', 'detection of pvc in XXXXX using the fractional XXXXX outperforms detection using XXXXX', ['ecg signals', 'linear prediction'])\n", "('method used for task', 'we propose an adaptive rv measure based fuzzy weighting XXXXX for XXXXX', ['subspace clustering', 'fmri data analysis'])\n", "('NONE', 'compared with the XXXXX , the normalized XXXXX can be used to locate the lesion in cortex , since the real cortex is far less-than-ideal model', ['gradient method', 'pca decomposition'])\n", "('NONE', 'a speckle reduction filter for XXXXX using XXXXX on coefficient of variation ( cv ) is proposed', ['ultrasound images', 'fuzzy logic'])\n", "('NONE', 'experiments based on well established database were carried out to evaluate the performance of the proposed approach . the results demonstrated better characterization of the movement assessment and motion recognition ability , with a XXXXX of 86.19 % , than the currently used methods such as XXXXX and pose normalization employed in motion recognition tasks', ['recognition rate', 'gaussian mixture models'])\n", "('NONE', 'the XXXXX is computationally efficient in the presence of high-resolution retinal XXXXX', ['fundus images', 'registration method'])\n", "('method used for task', 's-crc is tested for emg signal XXXXX for 10 XXXXX', ['pattern recognition', 'hand gestures'])\n", "('method used for task', 'in a former paper which is accepted in this journal ( which is under revision ) we implemented several filtering methods for hemodynamic XXXXX . there we showed in detail that the hemodynamic model is weakly non-linear . based on that finding we extended our researched by implementing a successful joint hemodynamic state and XXXXX . we also implemented the algorithm to a real fmri data set', ['state estimation', 'parameter estimation method'])\n", "('method used for task', 'in a former paper which is accepted in this journal ( which is under revision ) we implemented several filtering methods for hemodynamic XXXXX . there we showed in detail that the hemodynamic model is weakly non-linear . based on that finding we extended our researched by implementing a successful joint hemodynamic state and parameter estimation method . we also implemented the algorithm to a real XXXXX set', ['state estimation', 'fmri data'])\n", "('method used for task', 'in a former paper which is accepted in this journal ( which is under revision ) we implemented several filtering methods for hemodynamic state estimation . there we showed in detail that the hemodynamic model is weakly non-linear . based on that finding we extended our researched by implementing a successful joint hemodynamic state and XXXXX . we also implemented the algorithm to a real XXXXX set', ['parameter estimation method', 'fmri data'])\n", "('method used for task', 'we emphasized the following point in the literature . it is alleged that XXXXX is unrobust and poor in performance . we carefully examined this allegation under various state/measurement noise conditions . this allegation is not true for the widely used hemodynamic model in a wide range of process and measurement noise conditions . furthermore , several times XXXXX are alleged to outperform extended kalman filters . the alleged ekf is shown to be better in state estimation . we examined carefully and showed the contrary is true', ['extended kalman filter', 'particle filters'])\n", "('method used for task', 'we emphasized the following point in the literature . it is alleged that XXXXX is unrobust and poor in performance . we carefully examined this allegation under various state/measurement noise conditions . this allegation is not true for the widely used hemodynamic model in a wide range of process and measurement noise conditions . furthermore , several times particle filters are alleged to outperform extended kalman filters . the alleged ekf is shown to be better in XXXXX . we examined carefully and showed the contrary is true', ['extended kalman filter', 'state estimation'])\n", "('method used for task', 'we emphasized the following point in the literature . it is alleged that extended kalman filter is unrobust and poor in performance . we carefully examined this allegation under various state/measurement noise conditions . this allegation is not true for the widely used hemodynamic model in a wide range of process and measurement noise conditions . furthermore , several times XXXXX are alleged to outperform extended kalman filters . the alleged ekf is shown to be better in XXXXX . we examined carefully and showed the contrary is true', ['particle filters', 'state estimation'])\n", "('method used for task', 'the XXXXX is applied for the joint parameter and XXXXX . in the standard implementation of smoother there is no iteration . we compared iterative performance and noted substantial improvement in performance', ['iterative scheme', 'state estimation'])\n", "('NONE', '1/enhl negatively correlates with the XXXXX α of a de-trended XXXXX', ['scaling factor', 'fluctuation analysis'])\n", "('NONE', 'enhl and the XXXXX α reflect information about XXXXX of balance', ['scaling factor', 'motor control'])\n", "('method used for task', 'random forest , random tree , mlp network and XXXXX ( svm ) are employed for XXXXX', ['support vector machines', 'data classification'])\n", "('method used for task', 'random forest , XXXXX , mlp network and XXXXX ( svm ) are employed for data classification', ['support vector machines', 'random tree'])\n", "('method used for task', 'random forest , XXXXX , mlp network and support vector machines ( svm ) are employed for XXXXX', ['data classification', 'random tree'])\n", "('NONE', 'the XXXXX using the mit-bih arrhythmia database shows a high XXXXX of 97 %', ['performance evaluation', 'classification accuracy'])\n", "('NONE', 'the XXXXX using the mit-bih XXXXX shows a high classification accuracy of 97 %', ['performance evaluation', 'arrhythmia database'])\n", "('NONE', 'the performance evaluation using the mit-bih XXXXX shows a high XXXXX of 97 %', ['classification accuracy', 'arrhythmia database'])\n", "('NONE', 'the proposed work presents a novel framework for fast and fully XXXXX of od in XXXXX', ['automatic detection', 'fundus images'])\n", "('NONE', 'a new method for XXXXX of XXXXX is proposed', ['automatic segmentation', 'coronary arteries'])\n", "('method used for task', 'the high XXXXX achieved is represented by an area under XXXXX of az=0.961 with a training set of 40 images', ['detection rate', 'roc curve'])\n", "('NONE', 'an automatic multi-organ XXXXX of prostate XXXXX is proposed', ['magnetic resonance images', 'segmentation method'])\n", "('NONE', 'a 2d XXXXX is modeled using the inter-scale ratio of XXXXX', ['wavelet coefficients', 'gel image'])\n", "('NONE', 'the XXXXX obtained can help integrating a cgms into an XXXXX', ['artificial pancreas', 'error reduction'])\n", "('NONE', 'XXXXX of the XXXXX in frequency and time domain are derived', ['analytical solutions', 'dispersion model'])\n", "('NONE', 'XXXXX of the dispersion model in frequency and XXXXX are derived', ['analytical solutions', 'time domain'])\n", "('NONE', 'analytical solutions of the XXXXX in frequency and XXXXX are derived', ['dispersion model', 'time domain'])\n", "('method used for task', 'the work studies the XXXXX from wide band electrocardiogram ( ecg ) signals for classifying XXXXX ( mi ) stages', ['feature extraction', 'myocardial infarction'])\n", "('method used for task', 'we proposed a new efficient XXXXX for XXXXX', ['photoacoustic imaging', 'image reconstruction algorithm'])\n", "('NONE', 'the proposed XXXXX has better XXXXX to enter into long and thin indentations of medical images', ['snake model', 'convergence property'])\n", "('method used for task', 'the proposed XXXXX has better convergence property to enter into long and thin indentations of XXXXX', ['snake model', 'medical images'])\n", "('method used for task', 'the proposed snake model has better XXXXX to enter into long and thin indentations of XXXXX', ['convergence property', 'medical images'])\n", "('method used for task', 'warm XXXXX ( about 33°c ) should be maintained for reliable XXXXX', ['pulse oximetry', 'site temperature'])\n", "('method used for task', 'a compressive sampling pam system based on XXXXX ( cslrm-pam ) system has been set up to achieve fast XXXXX', ['data acquisition', 'low rank matrix completion'])\n", "('NONE', 'the final pam image is recovered directly from the compressive data by XXXXX without dramatic loss of XXXXX', ['low rank matrix completion', 'image resolution'])\n", "('NONE', 'XXXXX of XXXXX while running outdoors is deemed feasible', ['feedback control', 'heart rate'])\n", "('method used for task', 'on normal muscles , it performs better than other relevant algorithms , regarding XXXXX and quantitative XXXXX figures of merit', ['signal processing', 'muap waveform'])\n", "('method used for task', 'we review the XXXXX ( iqa ) for XXXXX', ['image quality assessment', 'medical images'])\n", "('NONE', 'majority of iqa was done on XXXXX ( mri ) , XXXXX ( ct ) , and ultrasonic images', ['magnetic resonance imaging', 'computed tomography'])\n", "('method used for task', 'the scheme was implemented in a XXXXX for XXXXX', ['software framework', 'mobile devices'])\n", "('NONE', 'the scale and shape parameters of the XXXXX are estimated using the XXXXX ( ml ) method', ['gamma distribution', 'maximum likelihood'])\n", "('NONE', 'a denoised estimate is made in the XXXXX by defining an XXXXX', ['input space', 'inverse mapping'])\n", "('NONE', 'XXXXX of 99.92 % and XXXXX of 99.73 %', ['classification accuracy', 'system reliability'])\n", "('NONE', 'XXXXX of XXXXX', ['search algorithm', 'circular code motifs'])\n", "('NONE', 'based on available XXXXX , the XXXXX is up to 90 %', ['experimental data', 'prediction accuracy'])\n", "('NONE', 'existing methods of pooling similar rare variants were based on quantities calculated from XXXXX . our method pools rare variants with similar XXXXX calculated based on the ratio of variant frequencies between cases and controls', ['relative risk', 'control group'])\n", "('NONE', 'we have used autodock vina to compare the XXXXX of potential inhibitors against an essential enzyme in XXXXX and its human homolog', ['plasmodium falciparum', 'binding affinities'])\n", "('method used for task', 'XXXXX optimizations need to be included in XXXXX', ['hydrogen bond network', 'simulation protocols'])\n", "('NONE', 'the role of XXXXX in complexes computed and biophysical property of XXXXX involved in complexes interface was considered', ['amino acids', 'water molecules'])\n", "('NONE', 'the XXXXX of XXXXX a was most significantly affected by the binding of the atp and peptide ligand', ['structure dynamics', 'protein kinase'])\n", "('NONE', 'XXXXX in the domains of the proteins show that repeats involved in the function of protein possess conserved XXXXX', ['sequence repeats', 'sequence motifs'])\n", "('method used for task', 'XXXXX programing is an effective method to evolve XXXXX ( odes ) model from observed time-series data', ['gene expression', 'ordinary differential equations'])\n", "('method used for task', 'XXXXX programing is an effective method to evolve ordinary differential equations ( odes ) model from observed XXXXX', ['gene expression', 'time-series data'])\n", "('NONE', 'gene expression programing is an effective method to evolve XXXXX ( odes ) model from observed XXXXX', ['ordinary differential equations', 'time-series data'])\n", "('method used for task', 'XXXXX for each vaccine modeled by XXXXX', ['tumor growth', 'artificial neural networks'])\n", "('NONE', 'we employed XXXXX , structural motif , XXXXX , and solvent accessibility calculation', ['secondary structure', 'conservation pattern'])\n", "('NONE', 'XXXXX revealed XXXXX of l-type pyruvate kinase n-domain in domains a and c of the protein', ['computational docking', 'binding sites'])\n", "('NONE', 'the algorithm basically utilizes XXXXX from XXXXX', ['phylogenetic tree', 'multiple sequence alignment'])\n", "('NONE', 'we propose a XXXXX for detection of XXXXX epistatic interaction in gwas', ['fast method', 'high order'])\n", "('NONE', 'a new XXXXX for chemicals using time-dependent cellular XXXXX is proposed', ['classification method', 'response profiles'])\n", "('NONE', 'a XXXXX machine-based XXXXX is proposed to predict conformational b-cell epitopes', ['support vector', 'ensemble method'])\n", "('method used for task', 'each gene pair was represented by incorporating the XXXXX and XXXXX', ['expression data', 'sequence information'])\n", "('method used for task', 'hierarchical closeness outperforms the other well-known structural XXXXX , particularly for cancer , hereditary , immune , and neurodegenerative disease-related genes in a human XXXXX', ['centrality measures', 'signaling network'])\n", "('NONE', 'establish our XXXXX basing on XXXXX', ['prediction model', 'support vector machine classifier'])\n", "('method used for task', 'we proposed a XXXXX to increase XXXXX', ['hybrid algorithm', 'metabolites production'])\n", "('NONE', 'we implement kinetic formalism of rene thomas along with XXXXX approaches to provide qualitative thresholds of entities in tlr3 associated biological regulatory network of XXXXX , satisfying the biological observations', ['model checking', 'dengue virus'])\n", "('NONE', 'different XXXXX of e296 ( d299 ) and d312 ( 315 ) in ompf and ompc promote the large-\\xadscale deviation in XXXXX and dynamics', ['protein structure', 'protonation states'])\n", "('method used for task', 'we proposed a XXXXX for protein–rna binding sites prediction by combining XXXXX and global features from protein sequence based on submodularity subset selection', ['computational method', 'local features'])\n", "('method used for task', 'we proposed a XXXXX for protein–rna binding sites prediction by combining local features and global features from XXXXX based on submodularity subset selection', ['computational method', 'protein sequence'])\n", "('method used for task', 'we proposed a computational method for protein–rna binding sites prediction by combining XXXXX and global features from XXXXX based on submodularity subset selection', ['local features', 'protein sequence'])\n", "('method used for task', 'proposed a novel basic XXXXX method used for ds XXXXX', ['evidence theory', 'probability assignment'])\n", "('NONE', 'we improved the XXXXX by adding a uniform XXXXX in the onlooker phase', ['abc algorithm', 'crossover operation'])\n", "('NONE', 'gnc is a XXXXX to the problem of XXXXX biological validation', ['robust solution', 'gene network'])\n", "('NONE', 'XXXXX of XXXXX were explored based on these degs', ['complex networks', 'hub genes'])\n", "('method used for task', 'XXXXX is accentuated by XXXXX', ['parkinson’s disease', 'insulin resistance'])\n", "('method used for task', 'we have incorporated these heuristics into a celebrated XXXXX called spici to get three new XXXXX', ['clustering algorithm', 'clustering algorithms'])\n", "('NONE', 'a method of identifying XXXXX based on moepga in XXXXX is proposed', ['protein complexes', 'ppi network'])\n", "('method used for task', 'here , XXXXX , dipeptide composition , correlation features , composition , transition and distribution and pseudo XXXXX are used to represent the XXXXX', ['amino acid composition', 'protein sequence'])\n", "('method used for task', 'here , XXXXX , dipeptide composition , correlation features , composition , transition and distribution and pseudo XXXXX are used to represent the XXXXX', ['amino acid composition', 'protein sequence'])\n", "('NONE', 'f1p and f6p can bind to the same XXXXX with different XXXXX', ['active site', 'binding mode'])\n", "('method used for task', 'two XXXXX were used to construct the optimized XXXXX', ['feature selection methods', 'feature subset'])\n", "('NONE', 'our method received a XXXXX using 10-fold cross-validation on the XXXXX', ['high performance', 'training dataset'])\n", "('method used for task', 'XXXXX and XXXXX conformational preferences', ['amino acid', 'secondary structure'])\n", "('method used for task', 'genes connected to the cd4+ t cell subtype XXXXX are more likely to be XXXXX', ['transcription factors', 'master genes'])\n", "('NONE', 'the XXXXX of increased succinate production after gnd gene knockout was consistent with XXXXX', ['experimental data', 'model prediction'])\n", "('method used for task', 'XXXXX for XXXXX can be done in a computationally efficient manner', ['parameter estimation', 'stochastic models'])\n", "('method used for task', 'the strategy combines XXXXX and XXXXX', ['computational modeling', 'experimental analysis'])\n", "('method used for task', 'XXXXX adjustment method ( psam ) is proposed as a tool for XXXXX to improve the power for single locus studies through an estimated XXXXX ( ps )', ['dimension reduction', 'propensity score'])\n", "('NONE', 'molecular interaction and XXXXX of hits were studied by XXXXX', ['binding mode', 'molecular docking'])\n", "('NONE', 'the XXXXX problem can be regarded as a special XXXXX problem in the sense that truly present proteins are those proteins with non-zero abundances', ['protein inference', 'protein quantification'])\n", "('method used for task', 'we test three very simple XXXXX methods to solve the XXXXX problem effectively', ['protein quantification', 'protein inference'])\n", "('method used for task', 'we used the computational strategy which includes XXXXX , secondary structure prediction , comparative modeling and XXXXX ( ppi ) analysis', ['sequence analysis', 'protein–protein interactions'])\n", "('NONE', 'XXXXX reach a stable XXXXX between tsp-1 domain and lskl peptide', ['md simulations', 'binding mode'])\n", "('NONE', 'a constrained XXXXX estimates probabilities in XXXXX', ['hidden markov model', 'linear time'])\n", "('NONE', 'we propose a novel algorithm for learning XXXXX with complete and XXXXX', ['factor analysis', 'incomplete data'])\n", "('method used for task', 'the XXXXX is the varying coefficient fractionally XXXXX', ['regression model', 'exponential model'])\n", "('method used for task', 'estimation of XXXXX for XXXXX is challenging', ['statistical models', 'social networks'])\n", "('method used for task', 'we use XXXXX to impute XXXXX enhancing survival tree analysis', ['bayesian networks', 'missing data'])\n", "('method used for task', 'the XXXXX is learned from XXXXX and used for the imputation', ['bayesian network', 'incomplete data'])\n", "('method used for task', 'a variational XXXXX is developed for XXXXX', ['em algorithm', 'parameter estimation'])\n", "('NONE', 'a new hbic criterion is proposed for XXXXX in XXXXX', ['model selection', 'mixture models'])\n", "('method used for task', 'we introduce apfa as XXXXX for discrete XXXXX', ['graphical models', 'longitudinal data'])\n", "('NONE', 'the lower and upper XXXXX are estimated by XXXXX', ['maximum likelihood', 'tail dependence parameters'])\n", "('method used for task', 'XXXXX and XXXXX show that the new approximations are very accurate for all practical purposes', ['theoretical analysis', 'empirical evaluation'])\n", "('method used for task', 'we derive value at risk and XXXXX for gtl-distributed XXXXX', ['expected shortfall', 'random variables'])\n", "('NONE', 'XXXXX show high XXXXX , low false positive rate', ['simulation studies', 'detection rate'])\n", "('NONE', 'XXXXX was demonstrated in simulations and real XXXXX', ['high performance', 'data analysis'])\n", "('NONE', 'under the pot framework , we estimate tail XXXXX . extensive XXXXX show the new method works well', ['risk measures', 'simulation studies'])\n", "('method used for task', 'XXXXX is constructed based on XXXXX to improve retrieval efficiency', ['case base', 'domain ontology'])\n", "('method used for task', 'the XXXXX is defined based on the XXXXX', ['similarity function', 'cost function'])\n", "('method used for task', 'automatic collection of constraint data from XXXXX for XXXXX', ['cad models', 'assembly planning'])\n", "('method used for task', 'generated XXXXX are validated via interference and XXXXX', ['stability analysis', 'assembly sequences'])\n", "('NONE', 'a schema is defined for XXXXX that represent XXXXX as networks', ['spatial relations', 'space layouts'])\n", "('method used for task', 'XXXXX for spatial XXXXX extend the schema', ['spatial constraints', 'consistency checking'])\n", "('NONE', 'XXXXX of XXXXX and simulation for milling in the field of dental technology', ['integrated approach', 'tool path generation'])\n", "('method used for task', 'a kind of XXXXX is adopted to describe the XXXXX', ['region codes', 'surface regions'])\n", "('NONE', 'we propose a comprehensive XXXXX geometrical XXXXX', ['human body', 'shape representation'])\n", "('NONE', 'XXXXX of recommended XXXXX types is implemented', ['automatic generation', 'assembly tolerance'])\n", "('NONE', 'perfect integration of XXXXX in a XXXXX', ['subdivision surfaces', 'cad system'])\n", "('NONE', 'a XXXXX classifies various designs as three-level XXXXX', ['representation model', 'design elements'])\n", "('method used for task', 'it combines csg modeling and XXXXX based shape and XXXXX', ['level set', 'topology optimization'])\n", "('method used for task', 'feedrate planning under confined XXXXX is reduced to XXXXX', ['tracking error', 'convex optimization'])\n", "('method used for task', 'XXXXX taubin integral for smooth or piecewise XXXXX', ['closed form', 'smooth surfaces'])\n", "('method used for task', 'lspia is so flexible that it allows the adjustment of the number of XXXXX , and a XXXXX in the iterations', ['control points', 'knot vector'])\n", "('NONE', 'numerical behavior of XXXXX through XXXXX', ['singular value decomposition', 'matrix representations'])\n", "('NONE', 'XXXXX of XXXXX and advection–diffusion demonstrated', ['isogeometric analysis', 'linear elasticity'])\n", "('NONE', 'algorithm for computing voxelized XXXXX of two XXXXX', ['minkowski sum', 'input polyhedra'])\n", "('method used for task', 'a new XXXXX for the XXXXX via geometric computations', ['analysis method', 'tolerance estimation'])\n", "('NONE', 'employing two pre-built XXXXX , a hierarchical gauss map and a coons bounding volume hierarchy , we develop an efficient culling technique that can eliminate the majority of redundant XXXXX', ['data structures', 'surface patches'])\n", "('NONE', 'demonstrated by a XXXXX of multiobjective XXXXX of a paper mill', ['case study', 'design optimization'])\n", "('method used for task', 'we model an XXXXX based method for mechanical XXXXX', ['ant colony optimization', 'assembly planning'])\n", "('method used for task', 'we propose a new approach to process eeg and XXXXX for XXXXX', ['hrv signals', 'design activities'])\n", "('NONE', 'family mould XXXXX using XXXXX and mould XXXXX grammars is proposed', ['genetic algorithms', 'layout design'])\n", "('NONE', 'group-oriented mould XXXXX grammar-based XXXXX and chromosome are proposed', ['layout design', 'genetic operators'])\n", "('NONE', 'novel use of the existing XXXXX that scale linearly with the XXXXX', ['data structures', 'problem size'])\n", "('NONE', 'optimal XXXXX of XXXXX with various geometric constraints is presented', ['degree reduction', 'bézier curves'])\n", "('NONE', 'optimal XXXXX of bézier curves with various XXXXX is presented', ['degree reduction', 'geometric constraints'])\n", "('NONE', 'optimal degree reduction of XXXXX with various XXXXX is presented', ['bézier curves', 'geometric constraints'])\n", "('NONE', 'XXXXX shapes are XXXXX representatives from a simulation perspective', ['skin model', 'skin shapes'])\n", "('NONE', 'XXXXX shapes are XXXXX representatives from a simulation perspective', ['skin shapes', 'skin model'])\n", "('method used for task', 'an efficient XXXXX suitable for XXXXX', ['interactive applications', 'parallel solver'])\n", "('method used for task', 'we present a novel bci based XXXXX for conceptual XXXXX', ['user interface', '3d modeling'])\n", "('NONE', 'a XXXXX study to assess intuitiveness of bci in XXXXX is presented', ['human factors', '3d modeling'])\n", "('method used for task', 'we present a method for automatic tool-adaptive XXXXX for XXXXX', ['path planning', 'freeform surfaces'])\n", "('method used for task', 'derivation of XXXXX for loaded XXXXX with a varying isotropic stress', ['shell structures', 'equilibrium equations'])\n", "('method used for task', 'the application of the XXXXX for the XXXXX', ['finite element method', 'numerical implementation'])\n", "('method used for task', 'the introduction of a new XXXXX for the XXXXX', ['particle method', 'numerical implementation'])\n", "('NONE', 'a multi-dimensional XXXXX based XXXXX fitting scheme to a freeform rational surface', ['dynamic programming', 'ruled surface'])\n", "('method used for task', 'highly XXXXX running on gpus are employed to evaluate the multi-dimensional XXXXX', ['parallel algorithms', 'dynamic programming'])\n", "('method used for task', 'unified XXXXX for both static–dynamic load analysis , XXXXX and form finding', ['structural optimization', 'physics engine'])\n", "('NONE', 'an overview is given that combine existing casting techniques with XXXXX for the fabrication of complex XXXXX', ['digital fabrication', 'concrete structures'])\n", "('NONE', 'present a perspective overview of the XXXXX about 3d XXXXX', ['future research', 'tolerance analysis'])\n", "('NONE', 'XXXXX were extracted based on a local score of XXXXX', ['boundary points', 'data points'])\n", "('NONE', 'we investigate the use of XXXXX for the exchange of “intelligent” XXXXX', ['semantic web technologies', 'cad models'])\n", "('method used for task', 'we define axioms and XXXXX to achieve XXXXX between cad ontologies', ['semantic integration', 'mapping rules'])\n", "('NONE', 'we extend XXXXX with a XXXXX to detect similar design features', ['semantic integration', 'similarity measurement'])\n", "('method used for task', 'we extend XXXXX with a similarity measurement to detect similar XXXXX', ['semantic integration', 'design features'])\n", "('method used for task', 'we extend semantic integration with a XXXXX to detect similar XXXXX', ['similarity measurement', 'design features'])\n", "('method used for task', 'the XXXXX is used to create a dual-level XXXXX . numerical instabilities can be improved', ['density approximation', 'shepard function'])\n", "('method used for task', 'a new spiral XXXXX based on space archimedean spiral is proposed to machining XXXXX with big slope', ['tool path generation', 'freeform surfaces'])\n", "('method used for task', 'we present shared XXXXX ( sst ) to rasterize XXXXX on gpus', ['implicit surfaces', 'subdivision trees'])\n", "('method used for task', 'XXXXX is adopted to train the XXXXX accumulatively', ['online learning', 'segmentation model'])\n", "('method used for task', 'we present a XXXXX to compute and update the mesh structure efficiently during XXXXX', ['parallel algorithm', 'image registration'])\n", "('method used for task', 'double tangential contact between the tool and the XXXXX is employed to connect feasible hyper-osculating XXXXX', ['tool paths', 'target surface'])\n", "('NONE', 'we introduce a XXXXX that maximizes the geometric matching between the tool and the XXXXX', ['global optimization algorithm', 'target surface'])\n", "('NONE', 'the algorithm analyzes high-level XXXXX existent in a XXXXX', ['semantic information', '3d model'])\n", "('NONE', 'we reduce the XXXXX dramatically by solving XXXXX', ['computation time', 'convex optimization problem'])\n", "('NONE', 'we propose two intrinsic methods for computing XXXXX ( cvt ) on XXXXX', ['centroidal voronoi tessellation', 'triangle meshes'])\n", "('method used for task', 'models are derived using XXXXX , design geometry and XXXXX', ['simulation intent', 'analysis attributes'])\n", "('NONE', 'by specifying a different XXXXX , different XXXXX are obtained', ['simulation intent', 'analysis models'])\n", "('NONE', 'XXXXX of surface meshes are introduced to XXXXX', ['geometric constraints', 'global optimization'])\n", "('NONE', 'XXXXX of XXXXX are introduced to global optimization', ['geometric constraints', 'surface meshes'])\n", "('method used for task', 'geometric constraints of XXXXX are introduced to XXXXX', ['global optimization', 'surface meshes'])\n", "('method used for task', 'a XXXXX is proposed to describe the local shape of the XXXXX', ['local descriptor', 'interest points'])\n", "('NONE', 'we prove that for a given XXXXX , the XXXXX exists only in very special cases ( for a special form of an angle function )', ['b-spline curve', 'exact solution'])\n", "('NONE', 'we propose an algorithm for finding an XXXXX , derive a bound on its XXXXX and study the approximation order of the proposed algorithm', ['approximate solution', 'approximation error'])\n", "('method used for task', 'XXXXX represents a crucial factor for good XXXXX', ['image quality', 'lens system design'])\n", "('NONE', 'XXXXX is the main part of the XXXXX methodology', ['optimization procedure', 'lens system design'])\n", "('NONE', 'XXXXX is modeled as a network of XXXXX agents', ['design space', 'design parameter'])\n", "('NONE', 'XXXXX agents help to overcome the cognitive limitation of XXXXX', ['design parameter', 'role agents'])\n", "('method used for task', 'a XXXXX is obtained using two different XXXXX', ['confidence interval', 'linearization strategies'])\n", "('NONE', 'an integrated XXXXX for identifying contextual impacts of XXXXX', ['product model', 'design changes'])\n", "('NONE', 'we propose a new approach to the problem of XXXXX of XXXXX', ['degree reduction', 'bézier curves'])\n", "('NONE', 'we devise a cooperative awareness model to describe cooperative XXXXX in XXXXX', ['product design', 'awareness information'])\n", "('method used for task', 'a mechanism which responds to changes of lean cooperative XXXXX is proposed to plan and execute task in XXXXX', ['product design', 'awareness information'])\n", "('method used for task', 'XXXXX has high geometric fidelity and low XXXXX', ['approximation error', 'steiner format'])\n", "('method used for task', 'an XXXXX to generate smooth spiral curves on XXXXX', ['efficient algorithm', 'free-form surfaces'])\n", "('method used for task', 'an XXXXX is given to improve the conformality of XXXXX', ['optimization method', 'nurbs surfaces'])\n", "('method used for task', 'we develop an analytical representation of XXXXX for XXXXX', ['conformal mapping', 'implicit surfaces'])\n", "('NONE', 'the XXXXX of the XXXXX is the description length', ['quality measure', 'statistical shape model'])\n", "('NONE', 'the XXXXX of the statistical shape model is the XXXXX', ['quality measure', 'description length'])\n", "('method used for task', 'the quality measure of the XXXXX is the XXXXX', ['statistical shape model', 'description length'])\n", "('method used for task', 'the approaches ground on algorithms from XXXXX and XXXXX', ['computational geometry', 'computer graphics'])\n", "('method used for task', 'we develop a symbolic XXXXX to construct trivariate solids and carry out XXXXX', ['modeling framework', 'isogeometric analysis'])\n", "('method used for task', 'a XXXXX has been developed for XXXXX of pmi in step files', ['software tool', 'conformance checking'])\n", "('NONE', 'a XXXXX has been developed for conformance checking of pmi in XXXXX', ['software tool', 'step files'])\n", "('NONE', 'a software tool has been developed for XXXXX of pmi in XXXXX', ['conformance checking', 'step files'])\n", "('NONE', 'a novel teeth XXXXX by combining optical scan data and dental XXXXX is introduced', ['modeling framework', 'ct images'])\n", "('method used for task', 'an XXXXX is developed to fill a XXXXX with an all-hex mesh', ['iterative algorithm', 'triangular mesh'])\n", "('NONE', 'examples of application of the strategy for XXXXX in XXXXX and cad', ['adaptive refinement', 'isogeometric analysis'])\n", "('method used for task', 'we propose a grid-free XXXXX for analytic XXXXX', ['geometric modeling', 'discretization scheme'])\n", "('NONE', 'we propose an efficient method for extracting XXXXX from the XXXXX', ['feature lines', 'shape collection'])\n", "('NONE', 'we solve the initial XXXXX on XXXXX by solving a first-order ode', ['triangle meshes', 'value problem'])\n", "('method used for task', 'we propose a method for fitting a g2 quadratic XXXXX to planar styling XXXXX', ['b-spline curve', 'design data'])\n", "('NONE', 'XXXXX is attained by using a non-uniform knot vector of the XXXXX', ['g2 continuity', 'b-spline curve'])\n", "('method used for task', 'XXXXX is attained by using a non-uniform XXXXX of the b-spline curve', ['g2 continuity', 'knot vector'])\n", "('NONE', 'g2 continuity is attained by using a non-uniform XXXXX of the XXXXX', ['b-spline curve', 'knot vector'])\n", "('method used for task', 'adjoint-based XXXXX for XXXXX', ['error estimates', 'euler equations'])\n", "('method used for task', 'an XXXXX for smoothing the XXXXX is proposed', ['iterative procedure', 'frame field'])\n", "('NONE', 'XXXXX using only XXXXX as inputs was developed', ['user interface', 'cad models'])\n", "('NONE', 'b-age derives a XXXXX or XXXXX only from salient 3d points', ['b-spline curve', 'surface estimation'])\n", "('method used for task', 'average strip width XXXXX and sample points selection method for 3+2-axis XXXXX', ['sculptured surface machining', 'estimation method'])\n", "('NONE', 'novel link between XXXXX ( discrete XXXXX ) and cagd ( smooth patchworks from bezier surfaces of degree ( 1,1 ) )', ['discrete differential geometry', 'affine minimal surfaces'])\n", "('method used for task', 'a XXXXX to discrete XXXXX , based on smooth patchworks', ['geometric approach', 'affine minimal surfaces'])\n", "('method used for task', 'a XXXXX to compute a μ-basis for complex XXXXX', ['fast algorithm', 'rational curves'])\n", "('method used for task', 'main properties and several applications from dynamics , XXXXX and geometry , XXXXX on polyhedra , motion of rigid body and positive definite matrices', ['medical imaging', 'curve design'])\n", "('NONE', 'main properties and several applications from dynamics , XXXXX and geometry , curve design on polyhedra , motion of XXXXX and positive definite matrices', ['medical imaging', 'rigid body'])\n", "('NONE', 'main properties and several applications from dynamics , medical imaging and geometry , XXXXX on polyhedra , motion of XXXXX and positive definite matrices', ['curve design', 'rigid body'])\n", "('method used for task', 'a novel XXXXX method is proposed to compute the intersection between a ray and a XXXXX using the XXXXX approximation', ['parametric surface', 'second order'])\n", "('method used for task', 'it is a geometric XXXXX which is less sensitive to XXXXX than newton–raphson and halley methods', ['initial conditions', 'iteration scheme'])\n", "('method used for task', 'modified t-splines are favorable both in adaptive XXXXX and XXXXX', ['geometric modeling', 'isogeometric analysis'])\n", "('method used for task', 'it is proven that linear XXXXX lead to reparameterized cubic XXXXX', ['quaternion polynomials', 'ph curves'])\n", "('NONE', 'spatial rational XXXXX of a class m=3,4,5,6 are derived in a XXXXX having 2m+4 degrees of freedom', ['closed form', 'ph curves'])\n", "('method used for task', 'our XXXXX are both monotonicity preserving and XXXXX', ['convexity preserving', 'limit functions'])\n", "('NONE', 'an XXXXX for bound of the XXXXX is designed', ['hausdorff distance', 'estimation method'])\n", "('NONE', 'efficient evaluation through a XXXXX of the XXXXX', ['linear transformation', 'control polygon'])\n", "('NONE', 'the scheme produces XXXXX spatial XXXXX in 3d', ['convexity preserving', 'limit curves'])\n", "('method used for task', 'the XXXXX are proved to be g1 continuous , while XXXXX show that they are also g2 smooth and fair', ['numerical examples', 'limit curves'])\n", "('NONE', 'several examples , involving the XXXXX parametric speed , XXXXX , tangent , normal , curvature , and offsets are presented', ['arc length', 'sweep curve'])\n", "('method used for task', 'the initial parameters of the nurbs-snake are obtained using a two step procedure in which firstly weights are obtained by solving a XXXXX and then XXXXX are evolved by solving a system of linear equation', ['control points', 'quadratic programming problem'])\n", "('NONE', 'the conditions for XXXXX between toric XXXXX are analyzed', ['g1 continuity', 'surface patches'])\n", "('method used for task', 'some practical XXXXX for XXXXX are developed', ['sufficient conditions', 'g1 continuity'])\n", "('NONE', 'the proposed methods allow existing XXXXX to fully exploit the advantageous properties of XXXXX within the context of prevailing cad geometry representations', ['cad systems', 'ph curves'])\n", "('NONE', 'new XXXXX of XXXXX', ['direct method', 'curve design'])\n", "('method used for task', 'we construct univariate XXXXX for XXXXX', ['subdivision schemes', 'noisy data'])\n", "('method used for task', 'a XXXXX is analyzed and validated by several XXXXX', ['statistical model', 'numerical examples'])\n", "('NONE', 'XXXXX of the algorithm adapts to the geometry of XXXXX', ['step size', 'parametric surface'])\n", "('method used for task', 'we derive a new XXXXX that is equal to the known XXXXX in most cases', ['upper bound', 'lower bound'])\n", "('method used for task', 'we devise a XXXXX for XXXXX based on rdf', ['data-driven approach', 'object modeling'])\n", "('NONE', 'we introduce XXXXX into XXXXX and co-segmentation', ['deep learning', '3d shape segmentation'])\n", "('NONE', 'the proposed scheme does not require any XXXXX of XXXXX', ['prior knowledge', 'network topology'])\n", "('NONE', 'ralp’s main disadvantage is its XXXXX as the message exchange rounds depend on the XXXXX . in this article , we introduce the concept of clusters , where the data transfer can proceed after a certain number of message exchange rounds', ['convergence time', 'neighborhood graph'])\n", "('NONE', 'ralp’s main disadvantage is its XXXXX as the message exchange rounds depend on the neighborhood graph . in this article , we introduce the concept of clusters , where the XXXXX can proceed after a certain number of message exchange rounds', ['convergence time', 'data transfer'])\n", "('NONE', 'ralp’s main disadvantage is its XXXXX as the XXXXX rounds depend on the neighborhood graph . in this article , we introduce the concept of clusters , where the data transfer can proceed after a certain number of XXXXX rounds', ['convergence time', 'message exchange'])\n", "('NONE', 'ralp’s main disadvantage is its XXXXX as the XXXXX rounds depend on the neighborhood graph . in this article , we introduce the concept of clusters , where the data transfer can proceed after a certain number of XXXXX rounds', ['convergence time', 'message exchange'])\n", "('NONE', 'ralp’s main disadvantage is its convergence time as the message exchange rounds depend on the XXXXX . in this article , we introduce the concept of clusters , where the XXXXX can proceed after a certain number of message exchange rounds', ['neighborhood graph', 'data transfer'])\n", "('method used for task', 'ralp’s main disadvantage is its convergence time as the XXXXX rounds depend on the XXXXX . in this article , we introduce the concept of clusters , where the data transfer can proceed after a certain number of XXXXX rounds', ['neighborhood graph', 'message exchange'])\n", "('method used for task', 'ralp’s main disadvantage is its convergence time as the XXXXX rounds depend on the XXXXX . in this article , we introduce the concept of clusters , where the data transfer can proceed after a certain number of XXXXX rounds', ['neighborhood graph', 'message exchange'])\n", "('NONE', 'ralp’s main disadvantage is its convergence time as the XXXXX rounds depend on the neighborhood graph . in this article , we introduce the concept of clusters , where the XXXXX can proceed after a certain number of XXXXX rounds', ['data transfer', 'message exchange'])\n", "('NONE', 'ralp’s main disadvantage is its convergence time as the XXXXX rounds depend on the neighborhood graph . in this article , we introduce the concept of clusters , where the XXXXX can proceed after a certain number of XXXXX rounds', ['data transfer', 'message exchange'])\n", "('NONE', 'proposing a new approach to XXXXX in XXXXX', ['barrier coverage', 'wireless sensor network'])\n", "('NONE', 'cause-effect relationships between service qos metrics and vn XXXXX being modelled as XXXXX', ['bayesian network', 'stress states'])\n", "('method used for task', 'concise and up-to-date review of XXXXX for XXXXX', ['quality assessment', 'video streaming services'])\n", "('method used for task', 'discussion of XXXXX and challenges for qoe in XXXXX', ['future trends', 'video streaming services'])\n", "('NONE', 'we present an integrated platform that considers XXXXX of XXXXX', ['social aspects', 'mobile sensing'])\n", "('method used for task', 'we prove the np-hardness of the problem and develop a XXXXX to minimize the XXXXX', ['heuristic algorithm', 'energy consumption'])\n", "('NONE', 'this finding allowed for positive verification of the social XXXXX in XXXXX', ['online communities', 'exchange theory'])\n", "('method used for task', 'we show how twitter XXXXX relate to XXXXX pricing', ['volume spikes', 'stock options'])\n", "('method used for task', 'we provide a XXXXX to solve policy-aware XXXXX', ['heuristic algorithm', 'peering problem'])\n", "('NONE', 'we examine the XXXXX of peering connections to cover XXXXX in today’s internet', ['lower bound', 'end users'])\n", "('method used for task', 'large-scale computational generation and motif XXXXX for the synthetic XXXXX . we obtain a distinct motif pattern for each such class', ['distribution analysis', 'topology classes'])\n", "('NONE', 'comprehensive XXXXX of XXXXX ( facebook , twitter , google plus ) from which we obtain three quantifiable characteristic motif fingerprints', ['online social networks', 'motif analysis'])\n", "('method used for task', 'we propose robust access control framework for a network which has allowed XXXXX to be connected to the internal network in order to enable seamless XXXXX', ['smart devices', 'data sharing'])\n", "('method used for task', 'smart device’s XXXXX such as location , app usage pattern , unlock failures are being considered for XXXXX and data confidentiality', ['sensor data', 'access control'])\n", "('method used for task', 'smart device’s XXXXX such as location , app usage pattern , unlock failures are being considered for access control and XXXXX', ['sensor data', 'data confidentiality'])\n", "('method used for task', 'smart device’s sensor data such as location , app usage pattern , unlock failures are being considered for XXXXX and XXXXX', ['access control', 'data confidentiality'])\n", "('method used for task', 'the algorithm supports both the XXXXX and XXXXX simultaneously', ['access control', 'data confidentiality'])\n", "('NONE', 'XXXXX are effectiveness , efficiency , XXXXX , and fairness', ['performance metrics', 'power consumption'])\n", "('NONE', 'to cope with large-scale XXXXX , we introduce techniques to make the sfmap framework scalable . we validate the effectiveness of the approach using large-scale XXXXX collected at a gateway point of internet access links', ['measurement data', 'traffic data'])\n", "('method used for task', 'XXXXX are a good choice for their XXXXX and low energy consumption', ['decision trees', 'high accuracy'])\n", "('NONE', 'barometers offer XXXXX , XXXXX and independence from position', ['high accuracy', 'energy efficiency'])\n", "('method used for task', 'a XXXXX based on henkin models is given for a predicative polymorphic calculus of XXXXX', ['denotational semantics', 'access control'])\n", "('NONE', 'several XXXXX included an extension of aspectj that supports the new XXXXX', ['case studies', 'annotation model'])\n", "('method used for task', 'presents two XXXXX for a+i programs and proves that both are fully abstract w.r.t . the XXXXX', ['trace semantics', 'operational semantics'])\n", "('NONE', 'we discuss a XXXXX of the moldable XXXXX', ['prototype implementation', 'debugger model'])\n", "('NONE', 'we establish a XXXXX that captures the XXXXX of language workbenches', ['feature model', 'design space'])\n", "('NONE', 'tao is the first attempt to apply the methodology of XXXXX in test and XXXXX', ['denotational semantics', 'oracle generation'])\n", "('method used for task', 'we analyze the problem and XXXXX for XXXXX in c++', ['design space', 'actor programming'])\n", "('method used for task', 'the XXXXX is an extension to the communicating event-loop XXXXX', ['domain model', 'actor model'])\n", "('NONE', 'the XXXXX retains the safety and liveness properties of the XXXXX', ['domain model', 'actor model'])\n", "('NONE', 'we provide an XXXXX and validation of the XXXXX', ['operational semantics', 'domain model'])\n", "('NONE', 'mcerlang is used for XXXXX of timed XXXXX based on monitors', ['model checking', 'rebeca models'])\n", "('method used for task', 'the existing XXXXX is extended to calculate the XXXXX for simulation results', ['simulation tool', 'confidence interval'])\n", "('method used for task', 'bayesian belief networks ( bbns ) and XXXXX ( fcms ) , as dynamic influence graphs , were applied to handle the task of medical knowledge formalization for XXXXX', ['fuzzy cognitive maps', 'decision support'])\n", "('NONE', 'for reasoning on these knowledge models , a XXXXX reasoning engine , eye , with the necessary plug-ins was developed in the XXXXX', ['general purpose', 'semantic web'])\n", "('method used for task', 'developing a XXXXX for constructing a network from a XXXXX in linear time', ['fast algorithm', 'time series'])\n", "('method used for task', 'developing a XXXXX for constructing a network from a time series in XXXXX', ['fast algorithm', 'linear time'])\n", "('NONE', 'developing a fast algorithm for constructing a network from a XXXXX in XXXXX', ['time series', 'linear time'])\n", "('NONE', 'a simulated XXXXX was used to elicit different levels of XXXXX', ['mental workload', 'process control task'])\n", "('method used for task', 'development of systematic XXXXX and knowledge aggregation methodology based on XXXXX', ['knowledge acquisition', 'staff interviews'])\n", "('method used for task', 'use of rad based XXXXX for representing process knowledge aggregated from XXXXX', ['process models', 'staff interviews'])\n", "('NONE', 'the proposed class of flexible weight functions allows emphasis on different XXXXX when comparing XXXXX between two groups', ['time periods', 'cumulative incidence functions'])\n", "('method used for task', 'a large improvement in power was shown when an early XXXXX was used to compare two XXXXX with an early difference', ['weight function', 'cumulative incidence functions'])\n", "('method used for task', 'the methodology can impose constraints on all different compartments , even if XXXXX measurements are not possible by augmenting the XXXXX with an observer', ['control system', 'drug concentration'])\n", "('NONE', 'it is possible to assess potential XXXXX in healthy subjects using XXXXX', ['cardiovascular risk', 'cluster analysis'])\n", "('method used for task', 'a fully automatic XXXXX is proposed for regional classification of the left ventricular wall in XXXXX of small animals', ['ultrasound images', 'processing pipeline'])\n", "('method used for task', 'the pipeline is implemented using state-of-the-art methods from XXXXX and XXXXX', ['computer vision', 'pattern classification'])\n", "('method used for task', 'good performance of the XXXXX is demonstrated on ultrasound data of living mice before and after artificially induced XXXXX', ['myocardial infarction', 'processing pipeline'])\n", "('method used for task', 'we achieve notable improvement in sperm XXXXX and fewer XXXXX compared to the state-of-the-art method', ['false positives', 'head detection'])\n", "('NONE', 'we design and built a XXXXX for the treatment of XXXXX based on asbru', ['decision support system', 'breast cancer'])\n", "('method used for task', 'the XXXXX is improved using XXXXX of vessel structure', ['computational efficiency', 'prior knowledge'])\n", "('NONE', 'we develop an integrated XXXXX ( called ihautisis ) for improving the work efficiency of XXXXX', ['surveillance system', 'infection control'])\n", "('NONE', 'we developed a method for XXXXX of fovea in XXXXX based on vessel-free zone and adaptive gaussian template', ['automated detection', 'fundus images'])\n", "('NONE', 'a semi-online XXXXX in the XXXXX is studied', ['patient scheduling problem', 'pathology laboratory'])\n", "('NONE', 'the approach reduces XXXXX of patients and improves XXXXX', ['waiting time', 'operations efficiency'])\n", "('method used for task', 'by applying XXXXX and support vector machine learning methods to separate pd from hc , we demonstrated 85 % XXXXX', ['classification accuracy', 'feature selection algorithms'])\n", "('method used for task', 'XXXXX based on handwriting analysis can be complementary method to diagnosis made by clinician or other XXXXX', ['decision support system', 'decision support systems'])\n", "('NONE', 'we used three XXXXX to improve the performance of the XXXXX', ['svm classifier', 'feature selection methods'])\n", "('NONE', '13 signal XXXXX were derived from segments of XXXXX', ['quality metrics', 'ecg waveforms'])\n", "('method used for task', 'an XXXXX for fully automatic XXXXX', ['embedded system', 'fall detection'])\n", "('NONE', 'a freely available database for evaluation of XXXXX , consisting of both accelerometric and XXXXX', ['fall detection', 'depth data'])\n", "('method used for task', 'our scheme is based on extended XXXXX and suitable for XXXXX in tmis', ['chaotic maps', 'data exchange'])\n", "('NONE', 'respiratory-gating of 18f-fdg XXXXX improves the accuracy of XXXXX', ['pet images', 'volume estimates'])\n", "('method used for task', 'reduction in XXXXX index ( bmi ) , body fat percentage , and XXXXX', ['body mass', 'body weight loss'])\n", "('NONE', 'XXXXX characteristics affect XXXXX', ['growth plate', 'soc development'])\n", "('NONE', 'the discriminant performance in this study , evaluated as az ( area under the XXXXX ) , for detecting the existence of endotracheal tubes was 0.943±0.009 , and the detection error of the XXXXX was 1.89±2.01mm', ['roc curve', 'tip location'])\n", "('NONE', 'couples the XXXXX of photon transport with XXXXX', ['monte carlo method', 'heat transfer'])\n", "('NONE', 'the new features can enhance the performance of XXXXX that discriminate between malignant and benign XXXXX', ['classification algorithms', 'skin lesions'])\n", "('NONE', 'validation on XXXXX as well as uva XXXXX', ['clinical data', 'simulation data'])\n", "('NONE', 'we propose valworkbench and describe its internal modules and classes , in order to provide the much needed XXXXX for the development and testing of new validation measures as well as XXXXX', ['software platform', 'clustering algorithms'])\n", "('NONE', 'the proposed model combines the demographic XXXXX with the findings of the initial screening mammogram to elicit the hidden and impeding risk of developing XXXXX', ['risk factors', 'breast cancer'])\n", "('method used for task', 'a novel scheme for extracting XXXXX based on morphological XXXXX ( mca ) algorithm is presented in this paper', ['retinal blood vessels', 'component analysis'])\n", "('method used for task', 'a parallel XXXXX ( pgica ) on gpu was proposed for XXXXX', ['group ica', 'fmri data analysis'])\n", "('method used for task', 'provides comprehensive genome scale analysis for users with XXXXX ( hpc ) and gene-level analysis for memory efficient XXXXX', ['high performance computing', 'personal computers'])\n", "('NONE', 'we proved the XXXXX of the XXXXX proposed', ['high accuracy', 'segmentation method'])\n", "('NONE', 'this paper presents an algorithm for registration of XXXXX using orthogonal XXXXX as features', ['retinal images', 'moment invariants'])\n", "('method used for task', 'the trs in test XXXXX with respect to reference XXXXX is estimated using similarity transformation', ['retinal image', 'test retinal image'])\n", "('method used for task', 'the test XXXXX is aligned with reference XXXXX using the estimated registration parameters', ['retinal image', 'test retinal image'])\n", "('method used for task', 'XXXXX was proposed for accurate XXXXX regardless of any variations in the environment loading', ['motion control', 'compensation method'])\n", "('NONE', 'we predicted the outcome of stroke using knowledge discovery process ( kdp ) methods , XXXXX ( ann ) and XXXXX ( svm ) models', ['artificial neural networks', 'support vector machine'])\n", "('method used for task', 'current XXXXX systems fail to transfer a XXXXX to the next available nurse , if the primary nurse is on the phone with another patient', ['nurse call', 'nurse systems'])\n", "('method used for task', 'current XXXXX systems fail to transfer a XXXXX to the next available nurse , if the primary nurse is on the phone with another patient', ['nurse systems', 'nurse call'])\n", "('method used for task', 'current XXXXX systems fail to transfer a XXXXX to the next available nurse , if the primary nurse is currently on another visit', ['nurse call', 'nurse systems'])\n", "('method used for task', 'current XXXXX systems fail to transfer a XXXXX to the next available nurse , if the primary nurse is currently on another visit', ['nurse systems', 'nurse call'])\n", "('method used for task', 'patient-specific XXXXX is necessary to obtain accurate XXXXX', ['cardiac output', 'wall shear stress'])\n", "('NONE', 'a system that provides a convenient way for physicians to retrieve and compare XXXXX among XXXXX providers about herniorrhaphy', ['clinical pathways', 'health care'])\n", "('NONE', 'a cross-correlation function and an average deviation between the continuous XXXXX and the interpolation of limited XXXXX samples are calculated', ['blood glucose', 'blood glucose samples'])\n", "('NONE', 'XXXXX ( emr ) can support a secure , real-time , point-of-care , patient centric XXXXX for clinical care', ['electronic medical record', 'information resource'])\n", "('NONE', '∼14 % lower fpf than XXXXX , bayesian , knn , XXXXX using the same two features', ['level set', 'svm classification'])\n", "('method used for task', 'wind-driven XXXXX ( wso ) is used for optimizing the XXXXX of a clinical decision support system', ['swarm optimization', 'rule base'])\n", "('method used for task', 'wind-driven XXXXX ( wso ) is used for optimizing the rule base of a XXXXX', ['swarm optimization', 'clinical decision support system'])\n", "('NONE', 'wind-driven swarm optimization ( wso ) is used for optimizing the XXXXX of a XXXXX', ['rule base', 'clinical decision support system'])\n", "('NONE', 'XXXXX are utilized as XXXXX', ['shape features', 'adjacency statistics'])\n", "('method used for task', 'we propose 3d XXXXX and tracking algorithm based on an XXXXX and a kalman filter', ['vessel segmentation', 'active contour model'])\n", "('method used for task', 'we propose 3d XXXXX and tracking algorithm based on an active contour model and a XXXXX', ['vessel segmentation', 'kalman filter'])\n", "('method used for task', 'we propose 3d vessel segmentation and tracking algorithm based on an XXXXX and a XXXXX', ['active contour model', 'kalman filter'])\n", "('method used for task', 'the proposed methodology concentrates its efforts on personalized/individual XXXXX including the family history and the demographic XXXXX to elicit the hidden risk of developing breast cancer', ['decision making', 'risk factors'])\n", "('method used for task', 'the proposed methodology concentrates its efforts on personalized/individual XXXXX including the family history and the demographic risk factors to elicit the hidden risk of developing XXXXX', ['decision making', 'breast cancer'])\n", "('NONE', 'the proposed methodology concentrates its efforts on personalized/individual XXXXX including the XXXXX and the demographic risk factors to elicit the hidden risk of developing breast cancer', ['decision making', 'family history'])\n", "('NONE', 'the proposed methodology concentrates its efforts on personalized/individual decision making including the family history and the demographic XXXXX to elicit the hidden risk of developing XXXXX', ['risk factors', 'breast cancer'])\n", "('method used for task', 'the proposed methodology concentrates its efforts on personalized/individual decision making including the XXXXX and the demographic XXXXX to elicit the hidden risk of developing breast cancer', ['risk factors', 'family history'])\n", "('method used for task', 'the proposed methodology concentrates its efforts on personalized/individual decision making including the XXXXX and the demographic risk factors to elicit the hidden risk of developing XXXXX', ['breast cancer', 'family history'])\n", "('method used for task', 'XXXXX on the simulated data to construct a XXXXX', ['predictive model', 'machine learning methods'])\n", "('method used for task', 'new XXXXX based on pso and outlier rejection with XXXXX', ['level set', 'image segmentation method'])\n", "('NONE', 'the use of XXXXX allows an optimal choice of XXXXX', ['pso algorithm', 'cluster centers'])\n", "('NONE', 'XXXXX based segmentation of XXXXX and cup is implemented', ['adaptive threshold', 'optic disc'])\n", "('NONE', 'investigating the XXXXX of long-term hrv for XXXXX in chf patients', ['discrimination power', 'risk assessment'])\n", "('method used for task', 'ecg , XXXXX , respiration , strides and gait XXXXX were measured', ['skin conductance', 'acceleration signals'])\n", "('NONE', 'XXXXX was combined with the kpa-weighted XXXXX as a feature to reduce the influence of noises far from tumor', ['spatial correlation', 'color information'])\n", "('method used for task', 'the proposed approach combined the neutrosophic XXXXX with level set algorithm for XXXXX', ['image segmentation', 'similarity score'])\n", "('NONE', 'neutrosophic similarity score is employed to deal with XXXXX on XXXXX', ['uncertain information', 'image segmentation'])\n", "('method used for task', 'neutrosophic XXXXX is employed to deal with XXXXX on image segmentation', ['uncertain information', 'similarity score'])\n", "('method used for task', 'neutrosophic XXXXX is employed to deal with uncertain information on XXXXX', ['image segmentation', 'similarity score'])\n", "('NONE', 'the results showed that we obtain nature enhanced XXXXX using our method and the XXXXX is very small', ['x-ray images', 'processing time'])\n", "('method used for task', 'XXXXX thermal contrast magnitude depends on the XXXXX and depth as well as on the breast density', ['steady state', 'tumour diameter'])\n", "('method used for task', 'transient thermal contrast trends present three important characteristics including the XXXXX , the transient peak and its corresponding XXXXX', ['response time', 'observation time'])\n", "('NONE', 'a stable XXXXX with suitable force and XXXXX is achieved', ['human–robot interaction', 'path tracking'])\n", "('method used for task', 'corroborating on the appliance of XXXXX for XXXXX', ['classification algorithms', 'image segmentation'])\n", "('method used for task', 'proposing a unified and fully XXXXX for both epicardial and mediastinal fats on cardiac XXXXX', ['ct images', 'automatic segmentation method'])\n", "('method used for task', 'a parameter-dependent analysis of XXXXX is developed through a specific XXXXX', ['articular cartilage', 'computational model'])\n", "('method used for task', 'this XXXXX establishes a basic guide the essential properties to be included into XXXXX of articular cartilage', ['parametric study', 'computational modeling'])\n", "('NONE', 'this XXXXX establishes a basic guide the essential properties to be included into computational modeling of XXXXX', ['parametric study', 'articular cartilage'])\n", "('NONE', 'this parametric study establishes a basic guide the essential properties to be included into XXXXX of XXXXX', ['computational modeling', 'articular cartilage'])\n", "('method used for task', 'this paper presents an XXXXX for the XXXXX of mammograms to assist radiologists in confirming their diagnosis in mammography screening', ['integrated system', 'automatic analysis'])\n", "('method used for task', 'we design a semantic cdss to enable data interoperability and XXXXX for patient-specific XXXXX', ['knowledge sharing', 'clinical decision support'])\n", "('method used for task', 'a constructive XXXXX ( cgp ) is proposed for XXXXX', ['genetic programming', 'epileptic seizure detection'])\n", "('NONE', 'we developed a multi-period XXXXX using the XXXXX', ['similarity measure', 'medical diagnosis method'])\n", "('NONE', 'robust performance without collecting XXXXX or retuning the XXXXX', ['experimental data', 'control parameters'])\n", "('NONE', 'optimize the XXXXX by a modified alternative XXXXX', ['objective function', 'iterative algorithm'])\n", "('method used for task', 'XXXXX is exploited for combining XXXXX to improve the performance', ['ensemble learning', 'multiple models'])\n", "('NONE', 'in this work glaucoma identification is done using XXXXX of XXXXX', ['optic disk', 'wavelet features'])\n", "('NONE', 'wavelet features are extracted from segmented and XXXXX removed XXXXX', ['blood vessels', 'optic disk'])\n", "('method used for task', 'XXXXX are extracted from segmented and XXXXX removed optic disk', ['blood vessels', 'wavelet features'])\n", "('method used for task', 'XXXXX are extracted from segmented and blood vessels removed XXXXX', ['optic disk', 'wavelet features'])\n", "('method used for task', 'several XXXXX are used for prominent XXXXX', ['machine learning algorithms', 'feature selection'])\n", "('method used for task', 'XXXXX is used to reduce the dimensionality of XXXXX', ['genetic algorithm', 'feature vector'])\n", "('method used for task', 'automated segmentation of XXXXX for skeletal XXXXX assessment', ['x-ray images', 'bone age'])\n", "('method used for task', 'a XXXXX was applied to track the endocardial XXXXX', ['non-rigid registration', 'boundary points'])\n", "('NONE', 'implement a single prospective XXXXX or XXXXX of phase i combination trials in oncology', ['clinical trial', 'simulation studies'])\n", "('NONE', 'a wireless XXXXX measuring device used together with a smart XXXXX was developed', ['blood pressure', 'mobile device'])\n", "('NONE', 'found the XXXXX of XXXXX for type ii diabetic patients', ['risk factors', 'liver cancer'])\n", "('method used for task', 'built a XXXXX for XXXXX prediction for diabetic patient', ['web-based application', 'liver cancer'])\n", "('method used for task', 'bp XXXXX ( bpann ) is improved by XXXXX ( ga ) and levenberg–marquardt ( lm ) algorithm to model the relation among principal components ( pcs ) and brain age', ['artificial neural network', 'hybrid genetic algorithm'])\n", "('method used for task', 'bp XXXXX ( bpann ) is improved by hybrid genetic algorithm ( ga ) and levenberg–marquardt ( lm ) algorithm to model the relation among XXXXX ( pcs ) and brain age', ['artificial neural network', 'principal components'])\n", "('method used for task', 'bp XXXXX ( bpann ) is improved by hybrid genetic algorithm ( ga ) and levenberg–marquardt ( lm ) algorithm to model the relation among principal components ( pcs ) and XXXXX', ['artificial neural network', 'brain age'])\n", "('method used for task', 'bp artificial neural network ( bpann ) is improved by XXXXX ( ga ) and levenberg–marquardt ( lm ) algorithm to model the relation among XXXXX ( pcs ) and brain age', ['hybrid genetic algorithm', 'principal components'])\n", "('method used for task', 'bp artificial neural network ( bpann ) is improved by XXXXX ( ga ) and levenberg–marquardt ( lm ) algorithm to model the relation among principal components ( pcs ) and XXXXX', ['hybrid genetic algorithm', 'brain age'])\n", "('method used for task', 'bp artificial neural network ( bpann ) is improved by hybrid genetic algorithm ( ga ) and levenberg–marquardt ( lm ) algorithm to model the relation among XXXXX ( pcs ) and XXXXX', ['principal components', 'brain age'])\n", "('NONE', 'analytical formulas for XXXXX , XXXXX , and flow rate are given', ['velocity profile', 'wall shear stress'])\n", "('method used for task', 'analytical formulas for XXXXX , wall shear stress , and XXXXX are given', ['velocity profile', 'flow rate'])\n", "('method used for task', 'analytical formulas for velocity profile , XXXXX , and XXXXX are given', ['wall shear stress', 'flow rate'])\n", "('NONE', 'a review about 3d XXXXX of pulmonary nodules in XXXXX is presented', ['automatic detection', 'ct images'])\n", "('NONE', 'in emergency and crisis situations , many XXXXX can be unserviceable because of damage to equipment or loss of power . thus , XXXXX over wireless communication to achieve uninterrupted network services is a major obstacle', ['data transmission', 'communication channels'])\n", "('method used for task', 'the proposed middleware was ported into an XXXXX , which is compatible with the actual network environment without the need for changing the original XXXXX', ['embedded system', 'system architecture'])\n", "('method used for task', 'the proposed middleware was ported into an XXXXX , which is compatible with the actual XXXXX without the need for changing the original system architecture', ['embedded system', 'network environment'])\n", "('method used for task', 'the proposed middleware was ported into an embedded system , which is compatible with the actual XXXXX without the need for changing the original XXXXX', ['system architecture', 'network environment'])\n", "('NONE', 'method focused on mass-preserving registration of XXXXX with XXXXX', ['large deformation', 'lung images'])\n", "('NONE', 'the work proposes a novel XXXXX that provides a set of methods for metabolomics and spectral XXXXX', ['r package', 'data analysis'])\n", "('NONE', 'the provided functions include data loading , pre-processing , metabolite identification , univariate/multivariate XXXXX , XXXXX and feature selection', ['data analysis', 'machine learning'])\n", "('method used for task', 'the provided functions include data loading , pre-processing , metabolite identification , univariate/multivariate XXXXX , machine learning and XXXXX', ['data analysis', 'feature selection'])\n", "('method used for task', 'the provided functions include data loading , pre-processing , metabolite identification , univariate/multivariate data analysis , XXXXX and XXXXX', ['machine learning', 'feature selection'])\n", "('NONE', 'the biomedical XXXXX aims to speed up the construction of biomedical domain-specific XXXXX', ['search engines', 'search engine framework'])\n", "('NONE', 'we made the implementation of hubness-aware XXXXX available in the pyhubs XXXXX', ['machine learning techniques', 'software package'])\n", "('NONE', 'XXXXX of systems with different XXXXX', ['comparative analysis', 'feature sets'])\n", "('NONE', 'understanding the XXXXX of the XXXXX', ['reliability analysis', 'cadx system'])\n", "('method used for task', 'four methods are tested : a perceptron multilayer , self-organising maps , a XXXXX and XXXXX', ['radial basis function neural network', 'decision trees'])\n", "('NONE', \"glomerulus diameter and bowman 's XXXXX in renal XXXXX indicate various diseases\", ['capsule thickness', 'microscopic images'])\n", "('method used for task', \"this work proposed the analysis particles algorithm based on median filter for morphological image processing to detect the renal corpuscle objects . afterwards , the XXXXX and bowman 's XXXXX are measured\", ['glomerulus diameter', 'capsule thickness'])\n", "('method used for task', 'we used the XXXXX model method on a XXXXX of the wrist joint', ['3d model', 'rigid body spring'])\n", "('NONE', 'we used the rigid body spring model method on a XXXXX of the XXXXX', ['3d model', 'wrist joint'])\n", "('NONE', 'we used the XXXXX model method on a 3d model of the XXXXX', ['rigid body spring', 'wrist joint'])\n", "('NONE', 'data obtained from earlier scans are used to construct a XXXXX with laplacian XXXXX', ['probabilistic atlas', 'mixture model'])\n", "('NONE', 'XXXXX obtained from a XXXXX is modeled for the ct image reconstruction', ['prior information', 'probabilistic atlas'])\n", "('NONE', 'mean , XXXXX , and root-mean-squared value parameters of the signal XXXXX are computed to characterize the signal fluctuations', ['standard deviation', 'envelope amplitude'])\n", "('method used for task', 'an epidemiologic system analysis of XXXXX is presented through a XXXXX', ['cardiovascular risk', 'bayesian network model'])\n", "('method used for task', 'the induced XXXXX was used to make inferences taking into account three reasoning patterns : XXXXX , evidential reasoning , and intercausal reasoning', ['bayesian network', 'causal reasoning'])\n", "('method used for task', 'XXXXX ( pe ) is a XXXXX to evaluate the irregularity of signals', ['permutation entropy', 'fast method'])\n", "('method used for task', 'a hybrid XXXXX proposed with svd and ezw for XXXXX', ['ecg signals', 'compression method'])\n", "('NONE', 'a framework including XXXXX , XXXXX and annotation combination stages for a confidence based benchmark dataset for retinal image processing is proposed', ['data selection', 'task assignment'])\n", "('method used for task', 'a framework including XXXXX , task assignment and annotation combination stages for a confidence based XXXXX for retinal image processing is proposed', ['data selection', 'benchmark dataset'])\n", "('method used for task', 'a framework including data selection , XXXXX and annotation combination stages for a confidence based XXXXX for retinal image processing is proposed', ['task assignment', 'benchmark dataset'])\n", "('method used for task', 'the framework is used to build a confidence based XXXXX for XXXXX', ['benchmark dataset', 'cyst segmentation'])\n", "('NONE', 'the main objective of this study was to investigate the impact of implementing a computer-based order entry system without XXXXX on the number of radiographs ordered for patients seen in the XXXXX', ['clinical decision support', 'emergency department'])\n", "('NONE', 'the main objective of this study was to investigate the impact of implementing a computer-based XXXXX system without XXXXX on the number of radiographs ordered for patients seen in the emergency department', ['clinical decision support', 'order entry'])\n", "('NONE', 'the main objective of this study was to investigate the impact of implementing a computer-based XXXXX system without clinical decision support on the number of radiographs ordered for patients seen in the XXXXX', ['emergency department', 'order entry'])\n", "('NONE', 'our results show a decrease in the number of radiographs ordered after computer-based XXXXX system implementation , despite an increase in the number of XXXXX admissions', ['emergency department', 'order entry'])\n", "('NONE', 'our study also shows that the XXXXX between XXXXX admission and medical imaging was not affected by this new workflow', ['time interval', 'emergency department'])\n", "('method used for task', 'our study also shows that the XXXXX between emergency department admission and XXXXX was not affected by this new workflow', ['time interval', 'medical imaging'])\n", "('method used for task', 'our study also shows that the time interval between XXXXX admission and XXXXX was not affected by this new workflow', ['emergency department', 'medical imaging'])\n", "('method used for task', 'our proposed XXXXX score is similar to the XXXXX of glaucoma experts', ['quality assessment', 'quality index'])\n", "('NONE', 'a XXXXX of XXXXX ( tf ) was proposed by a tlm model of human artery tree', ['transfer function', 'calculation method'])\n", "('method used for task', 'this study proposed and examined a new approach employing the light sharing pet XXXXX with thick light guide and gapd array having large-area microcells for high effective XXXXX', ['detector configuration', 'quantum efficiency'])\n", "('method used for task', 'ensemble XXXXX and rrelief algorithm were applied to establish the XXXXX', ['machine learning', 'quantitative assessment model'])\n", "('NONE', 'we use inverse probability of treatment weighting , a XXXXX based technique , for covariate adjustment of the cumulative incidence functions in competing XXXXX', ['propensity score', 'risk analysis'])\n", "('NONE', 'we use inverse probability of treatment weighting , a XXXXX based technique , for covariate adjustment of the XXXXX in competing risk analysis', ['propensity score', 'cumulative incidence functions'])\n", "('NONE', 'we use inverse probability of treatment weighting , a propensity score based technique , for covariate adjustment of the XXXXX in competing XXXXX', ['risk analysis', 'cumulative incidence functions'])\n", "('method used for task', 'XXXXX ( pca ) for dominant XXXXX', ['principal component analysis', 'feature selection'])\n", "('NONE', 'the XXXXX presented allows the deployment of different benchmarking tests for XXXXX', ['software tool', 'm2m protocols'])\n", "('NONE', 'the most relevant XXXXX were evaluated considering different specific XXXXX', ['performance metrics', 'm2m protocols'])\n", "('method used for task', 'the peristaltic flow of a copper oxide water fluid investigate the effects of XXXXX and XXXXX', ['heat generation', 'magnetic field'])\n", "('NONE', 'we have developed a freely available XXXXX for semi-automated tracking of muscle fascicles in b-mode XXXXX', ['software package', 'ultrasound image sequences'])\n", "('method used for task', 'XXXXX and a XXXXX of the lumbar spine allow the prediction of fracture probability', ['evolutionary algorithm', 'finite element model'])\n", "('NONE', 'XXXXX and a finite element model of the XXXXX allow the prediction of fracture probability', ['evolutionary algorithm', 'lumbar spine'])\n", "('NONE', 'XXXXX and a finite element model of the lumbar spine allow the prediction of XXXXX', ['evolutionary algorithm', 'fracture probability'])\n", "('NONE', 'evolutionary algorithm and a XXXXX of the XXXXX allow the prediction of fracture probability', ['finite element model', 'lumbar spine'])\n", "('NONE', 'evolutionary algorithm and a XXXXX of the lumbar spine allow the prediction of XXXXX', ['finite element model', 'fracture probability'])\n", "('NONE', 'evolutionary algorithm and a finite element model of the XXXXX allow the prediction of XXXXX', ['lumbar spine', 'fracture probability'])\n", "('NONE', 'one of the interesting parts of the study is the XXXXX that significantly affects XXXXX', ['parameter optimization', 'system performance'])\n", "('NONE', 'we have developed a XXXXX and an algorithm that allow XXXXX of abnormal values of vital parameters occurring during anesthesia', ['data model', 'automatic detection'])\n", "('NONE', 'we propose an XXXXX combining semi XXXXX ( ssl ) and active learning ( al ) for segmenting crohns disease affected regions in mri', ['interactive method', 'supervised learning'])\n", "('method used for task', 'we propose an XXXXX combining semi supervised learning ( ssl ) and XXXXX ( al ) for segmenting crohns disease affected regions in mri', ['interactive method', 'active learning'])\n", "('method used for task', 'we propose an interactive method combining semi XXXXX ( ssl ) and XXXXX ( al ) for segmenting crohns disease affected regions in mri', ['supervised learning', 'active learning'])\n", "('NONE', 'compared to fully supervised methods we obtain high XXXXX with fewer samples and lesser XXXXX', ['computation time', 'segmentation accuracy'])\n", "('NONE', 'we have created a software poped lite in order to increase the use of XXXXX in preclinical XXXXX', ['optimal design', 'drug discovery'])\n", "('method used for task', 'the proposed approach uses XXXXX to XXXXX', ['genetic algorithms', 'dimensionality reduction'])\n", "('method used for task', 'XXXXX and XXXXX are used to recognize the signals', ['k-nearest neighbor', 'naive bayes classifiers'])\n", "('NONE', 'the system achieved a high XXXXX and can be implemented in an XXXXX', ['classification accuracy', 'embedded system'])\n", "('NONE', 'the feature extraction method combines XXXXX , XXXXX and fisher distance', ['wavelet packet decomposition', 'common spatial patterns'])\n", "('NONE', 'the XXXXX combines XXXXX , common spatial patterns and fisher distance', ['wavelet packet decomposition', 'feature extraction method'])\n", "('NONE', 'the XXXXX combines wavelet packet decomposition , XXXXX and fisher distance', ['common spatial patterns', 'feature extraction method'])\n", "('NONE', 'an up-to-date review about the proposed techniques for the XXXXX of pigmented XXXXX is presented', ['image segmentation', 'skin lesions'])\n", "('NONE', 'investigating application of cnt in improving XXXXX of XXXXX', ['dynamic response', 'conical shell'])\n", "('method used for task', 'a rotation-based isogeometric reissner–mindlin shell formulation which is able to handle XXXXX and XXXXX is presented', ['finite rotations', 'large deformations'])\n", "('method used for task', 'a XXXXX phase-field approach to 3d XXXXX', ['higher order', 'brittle fracture'])\n", "('NONE', 'domain with singularities and/or XXXXX discretised into XXXXX', ['complex geometry', 'finite elements'])\n", "('NONE', 'the dual form of XXXXX , abbreviated as dda-d , with the XXXXX as the basic variables , is established', ['discontinuous deformation analysis', 'contact forces'])\n", "('method used for task', 'bézier extraction of truncated XXXXX and the application of the approach to adaptive XXXXX are presented', ['hierarchical b-splines', 'isogeometric analysis'])\n", "('NONE', 'a strict element viewpoint is adopted in the XXXXX which facilitates the application of standard procedures of XXXXX', ['refinement algorithms', 'adaptive finite element analysis'])\n", "('NONE', 'XXXXX demonstrate optimal XXXXX for problems that involve singularities and strong gradients', ['numerical examples', 'convergence rates'])\n", "('NONE', 'first XXXXX of the fractional XXXXX', ['numerical solution', 'navier–stokes equations'])\n", "('method used for task', 'stress XXXXX for contacting cracks are validated against XXXXX', ['analytical solutions', 'intensity factors'])\n", "('NONE', 'maximum entropy XXXXX handle dispersion errors better than XXXXX', ['basis functions', 'finite elements'])\n", "('NONE', 'explicit integral form of the XXXXX facilitates XXXXX', ['sensitivity analysis', 'distance constraint'])\n", "('method used for task', 'a subjective XXXXX on the XXXXX of http adaptive streaming ( has ) is conducted', ['user study', 'influence factors'])\n", "('method used for task', 'we model a cross-layer problem of XXXXX , link delay , and XXXXX', ['congestion control', 'power control'])\n", "('NONE', 'we simplify the auction procedure by allowing the users to skip the XXXXX when the user is unsure or unaware of the exact XXXXX of his own repeated jobs', ['utility function', 'utility model'])\n", "('NONE', 'we implement abacus by modifying the XXXXX in hadoop , and test it on a large-scale cloud platform . our experimental results verify the truthfulness of our auction-based mechanism , XXXXX , as well as the accuracy of our utility prediction algorithm', ['scheduling algorithm', 'system efficiency'])\n", "('NONE', 'we support the selection of a XXXXX depending on XXXXX', ['user requirements', 'modeling technique'])\n", "('method used for task', 'we study the possibility to employ XXXXX ( mt ) systems and supervised methods for multilingual XXXXX', ['machine translation', 'sentiment analysis'])\n", "('NONE', 'unlike traditional hybrid XXXXX , ours consists of two XXXXX', ['language models', 'independent components'])\n", "('NONE', 'global prosodic and laughter-specific XXXXX were extracted for the two types of laughter . these parameters were analysed by XXXXX and classification trees to reduce the number of parameters', ['acoustic features', 'principal component analysis'])\n", "('method used for task', 'a XXXXX was trained and tested using seven important features , and total XXXXX was confirmed to be at least over 84 with unseen test material', ['support vector machine', 'classification accuracy'])\n", "('method used for task', 'XXXXX is affected by environmental and interlocutor-related XXXXX', ['speech production', 'contextual factors'])\n", "('NONE', 'the class-specific XXXXX scheme improves the XXXXX', ['multiple classifiers', 'recognition accuracy'])\n", "('NONE', 'XXXXX are usually implemented XXXXX and difficult to adapt to new domains', ['ad hoc', 'dialog managers'])\n", "('method used for task', 'XXXXX for spoken XXXXX ( std ) on english ( meeting domain ) and spanish ( read speech ) data in a discriminative confidence estimation framework', ['feature analysis', 'term detection'])\n", "('method used for task', 'XXXXX is based on groups that are defined according to their XXXXX : lattice-based features , duration-based features , lexical features , levenshtein distance-based features , position and prosodic features ( pitch and energy )', ['feature analysis', 'information sources'])\n", "('method used for task', 'XXXXX employs two well-known and established models : XXXXX ( a generative approach ) and logistic XXXXX ( a discriminative approach ) . individual and incremental analyses are presented for both models', ['feature analysis', 'linear regression'])\n", "('method used for task', 'XXXXX employs two well-known and established models : XXXXX ( a generative approach ) and logistic XXXXX ( a discriminative approach ) . individual and incremental analyses are presented for both models', ['feature analysis', 'linear regression'])\n", "('method used for task', 'the best XXXXX comprises features from different groups : lattice-based and lexical features are among the most informative groups in general , and duration and energy are more informative for read XXXXX', ['feature set', 'speech data'])\n", "('method used for task', 'XXXXX for a spanish-to-lse XXXXX for wide-domain application', ['rule-based approach', 'machine translation system'])\n", "('NONE', 'application of e-pca for automatic XXXXX in a XXXXX', ['spoken dialogue system', 'belief compression'])\n", "('method used for task', 'first real XXXXX comparing manual and automatic XXXXX', ['user evaluation', 'belief compression'])\n", "('NONE', 'acoustic XXXXX between spoken segments can improve spoken XXXXX', ['feature similarity', 'term detection'])\n", "('NONE', 'detailed XXXXX of results using multiple XXXXX', ['comparative analysis', 'evaluation metrics'])\n", "('NONE', 'we model the XXXXX by a speaker adapted XXXXX and a vocal tract filter', ['speech signal', 'glottal source'])\n", "('method used for task', 'introduce novel XXXXX for XXXXX', ['emotion recognition', 'ranking models'])\n", "('method used for task', 'simplified and supervised i-vector modeling for robust and efficient XXXXX and XXXXX', ['language identification', 'speaker verification'])\n", "('NONE', 'lexical feature with XXXXX using XXXXX between words and dimension as weights is powerful', ['sparse representation', 'mutual information'])\n", "('NONE', 'XXXXX with both XXXXX and channel distortion', ['speech enhancement', 'additive noise'])\n", "('method used for task', 'use of XXXXX to reduce training and testing data mismatch for XXXXX', ['speech recognition', 'corpus data'])\n", "('method used for task', 'improved results on aurora 4 for XXXXX and XXXXX', ['speech recognition', 'speech enhancement'])\n", "('method used for task', 'XXXXX and XXXXX are used for proximity computation', ['euclidean distance', 'cosine dissimilarity'])\n", "('NONE', 'XXXXX unlike XXXXX makes weighted pagerank to converge', ['cosine similarity', 'cosine dissimilarity'])\n", "('NONE', 'refinements based on XXXXX further reduce XXXXX', ['predictive models', 'segmentation errors'])\n", "('method used for task', 'we explain in detail the different steps in computing a XXXXX based on a XXXXX', ['language model', 'recurrent neural network'])\n", "('method used for task', 'word-level XXXXX for perplexity-based XXXXX', ['linguistic information', 'data selection'])\n", "('method used for task', 'for that we perform an XXXXX automatic post-editing from ready-to-use generic XXXXX', ['online learning', 'machine translation systems'])\n", "('method used for task', 'a XXXXX is developed for determining the XXXXX', ['model parameters', 'parameter estimation method'])\n", "('NONE', 'a new type of pole-zero XXXXX of the XXXXX is obtained', ['transfer function', 'vocal tract'])\n", "('NONE', 'XXXXX in a hri domain shows that the approach outperforms traditional hand-crafted and XXXXX', ['user evaluation', 'statistical models'])\n", "('NONE', 'XXXXX with XXXXX that represent similarity measures', ['distance functions', 'probabilistic semantics'])\n", "('NONE', 'XXXXX with probabilistic semantics that represent XXXXX', ['distance functions', 'similarity measures'])\n", "('NONE', 'distance functions with XXXXX that represent XXXXX', ['probabilistic semantics', 'similarity measures'])\n", "('NONE', 'the proposed XXXXX can estimate the direction of a signal in a XXXXX', ['noisy environment', 'doa methods'])\n", "('NONE', 'XXXXX per unit type based synthesis system generates high XXXXX', ['neural network', 'speech quality'])\n", "('NONE', 'integrating the XXXXX in XXXXX', ['principal component analysis', 'wavelet packet decomposition'])\n", "('method used for task', 'the purpose of this contribution is to review the state of the art in both areas , XXXXX and XXXXX', ['statistical methods', 'speech processing'])\n", "('NONE', 'borrowing theories from XXXXX , we propose a XXXXX of human cognition', ['cognitive psychology', 'computational model'])\n", "('NONE', 'borrowing theories from XXXXX , we propose a computational model of XXXXX', ['cognitive psychology', 'human cognition'])\n", "('NONE', 'borrowing theories from cognitive psychology , we propose a XXXXX of XXXXX', ['computational model', 'human cognition'])\n", "('method used for task', 'we verify the XXXXX and XXXXX with narrative text data', ['cognitive model', 'summarization method'])\n", "('NONE', 'preprocessed elderly XXXXX were tested with an android XXXXX', ['smart phone', 'voice signals'])\n", "('method used for task', 'XXXXX increased to 1.5 % by increasing the XXXXX', ['speech rate', 'speech recognition accuracy'])\n", "('method used for task', 'a new , particular XXXXX is proposed to improve XXXXX for automatic speech recognition', ['logistic regression model', 'confidence measures'])\n", "('method used for task', 'a new , particular XXXXX is proposed to improve confidence measures for XXXXX', ['logistic regression model', 'automatic speech recognition'])\n", "('method used for task', 'a new , particular logistic regression model is proposed to improve XXXXX for XXXXX', ['confidence measures', 'automatic speech recognition'])\n", "('method used for task', 'we develop the XXXXX based on XXXXX', ['voice activity detection', 'statistical model'])\n", "('NONE', 'XXXXX can be defined by pos tags and automatic XXXXX', ['word clustering', 'word classes'])\n", "('method used for task', 'XXXXX are used to filter synthetic utterances for one XXXXX', ['acoustic features', 'input concept'])\n", "('method used for task', 'this work presents a novel comprehensive study on the use of XXXXX ( dnns ) for automatic XXXXX ( lid )', ['deep neural networks', 'language identification'])\n", "('method used for task', 'best XXXXX is achieved by combining the results from the proposed bottleneck system and the baseline XXXXX . compared to a state-of-the-art XXXXX , the combined system achieves a 45 % of relative improvement in both eer and cavg , on 3s and 10s', ['system performance', 'i-vector system'])\n", "('NONE', 'best system performance is achieved by combining the results from the proposed bottleneck system and the baseline XXXXX . compared to a state-of-the-art XXXXX , the combined system achieves a 45 % of relative improvement in both eer and cavg , on 3s and 10s', ['baseline system', 'i-vector system'])\n", "('method used for task', 'this special issue is devoted to XXXXX and XXXXX', ['natural language processing', 'human–computer interaction'])\n", "('NONE', 'originality of the topic for XXXXX in XXXXX', ['knowledge improvement', 'pdca methodology'])\n", "('NONE', 'the proposed pim is proven feasible for XXXXX via several XXXXX', ['field studies', 'soap frameworks'])\n", "('method used for task', 'the design and implementation of jujube XXXXX and XXXXX', ['real-time monitoring', 'management system'])\n", "('NONE', 'XXXXX improve the XXXXX significantly in sparse networks', ['unidirectional links', 'network lifetime'])\n", "('NONE', 'reference framework for XXXXX and improvement in XXXXX', ['process assessment', 'systems engineering'])\n", "('method used for task', 'XXXXX is employed to extract customer utilities for generating XXXXX in a customer-driven approach', ['conjoint analysis', 'design concepts'])\n", "('method used for task', 'med-trace is a lightweight XXXXX and XXXXX', ['traceability assessment', 'improvement method'])\n", "('NONE', 'based on the XXXXX , the XXXXX is proposed', ['routing algorithm', 'address configuration algorithm'])\n", "('NONE', 'an integrated platform of XXXXX in an XXXXX is proposed', ['data exchange', 'intelligent transportation systems'])\n", "('method used for task', 'most apps are developed without applying XXXXX and XXXXX', ['information security', 'best practices'])\n", "('method used for task', 'XXXXX for XXXXX', ['situational awareness', 'critical infrastructure protection'])\n", "('method used for task', 'a generic XXXXX for XXXXX data', ['data model', 'food consumption'])\n", "('method used for task', 'we have used XXXXX to define XXXXX', ['similarity measures', 'similarity networks'])\n", "('method used for task', 'we have defined a XXXXX based on XXXXX in order to establish trust along a path of entities', ['trust model', 'similarity networks'])\n", "('NONE', 'a XXXXX evaluates the reduction of bus utilization and initialization times under XXXXX', ['simulation study', 'data compression'])\n", "('method used for task', 'this paper proposes a XXXXX for developing solutions for XXXXX', ['software architecture', 'smart environments'])\n", "('NONE', 'XXXXX involves two direct benefits : less time of developing , and XXXXX', ['software architecture', 'cost reduction'])\n", "('method used for task', 'the XXXXX based on XXXXX is proposed', ['network architecture', 'location information'])\n", "('method used for task', 'XXXXX and XXXXX based method are proposed', ['image histogram', 'human vision system'])\n", "('NONE', 'XXXXX should be measured taking into account XXXXX', ['color images', 'human vision system'])\n", "('method used for task', 'context-aware XXXXX for XXXXX planning ( erp )', ['classification methodology', 'enterprise resource'])\n", "('NONE', 'the XXXXX can also be applied in XXXXX', ['dynamic environment', 'discovery scheme'])\n", "('method used for task', 'modelling the XXXXX requires a way to convey the XXXXX', ['dynamic constraints', 'state transitions'])\n", "('method used for task', 'we define a XXXXX for XXXXX by extending square XXXXX', ['quality model', 'semantic technologies'])\n", "('method used for task', 'we define a XXXXX for XXXXX by extending square XXXXX', ['semantic technologies', 'quality model'])\n", "('NONE', 'this research is in the field of hardware based XXXXX . even though XXXXX are the base of this work but the algorithm has many other applications', ['image processing', 'microarray images'])\n", "('NONE', 'hardware compression of XXXXX is just an example of XXXXX', ['microarray images', 'big data applications'])\n", "('NONE', 'field programmable XXXXX based XXXXX module reduces conversion time', ['gate array', 'analog input'])\n", "('NONE', 'field programmable XXXXX based analog input module reduces XXXXX', ['gate array', 'conversion time'])\n", "('NONE', 'field programmable gate array based XXXXX module reduces XXXXX', ['analog input', 'conversion time'])\n", "('NONE', 'an lp framework that is capable of jointly modeling XXXXX and route diversity in XXXXX is developed', ['energy dissipation', 'wireless sensor networks'])\n", "('NONE', 'an XXXXX that is capable of jointly modeling XXXXX and route diversity in wireless sensor networks is developed', ['energy dissipation', 'lp framework'])\n", "('method used for task', 'an lp framework that is capable of jointly modeling XXXXX and XXXXX in wireless sensor networks is developed', ['energy dissipation', 'route diversity'])\n", "('NONE', 'an XXXXX that is capable of jointly modeling energy dissipation and route diversity in XXXXX is developed', ['wireless sensor networks', 'lp framework'])\n", "('NONE', 'an lp framework that is capable of jointly modeling energy dissipation and XXXXX in XXXXX is developed', ['wireless sensor networks', 'route diversity'])\n", "('method used for task', 'an XXXXX that is capable of jointly modeling energy dissipation and XXXXX in wireless sensor networks is developed', ['lp framework', 'route diversity'])\n", "('method used for task', 'lp problems are solvable in XXXXX . hence , from computational point of view it is preferable to model a problem using lp rather than with a nonpolynomial time formulation which makes a comprehensive analysis infeasible . thus , the novel XXXXX presented in this paper can be used with minor modifications for future analysis on different aspects of route diversity', ['polynomial time', 'lp framework'])\n", "('method used for task', 'lp problems are solvable in XXXXX . hence , from computational point of view it is preferable to model a problem using lp rather than with a nonpolynomial time formulation which makes a comprehensive analysis infeasible . thus , the novel lp framework presented in this paper can be used with minor modifications for future analysis on different aspects of XXXXX', ['polynomial time', 'route diversity'])\n", "('NONE', 'lp problems are solvable in polynomial time . hence , from computational point of view it is preferable to model a problem using lp rather than with a nonpolynomial time formulation which makes a comprehensive analysis infeasible . thus , the novel XXXXX presented in this paper can be used with minor modifications for future analysis on different aspects of XXXXX', ['lp framework', 'route diversity'])\n", "('NONE', 'we propose schemes to ensure prioritization and reduce XXXXX in XXXXX', ['end-to-end delay', 'hybrid network'])\n", "('NONE', 'use the XXXXX generated by the XXXXX to generate a videogame', ['process models', 'domain experts'])\n", "('method used for task', 'XXXXX based on tripartite credibility for XXXXX are proposed', ['rfid systems', 'secure mechanisms'])\n", "('NONE', 'the XXXXX have advantages in communication and XXXXX', ['computational cost', 'secure mechanisms'])\n", "('method used for task', 'a new convolutional XXXXX with three convolutional layers and three fully-connected layers is introduced . XXXXX is utilized to model the output error of previous networks and adjust configurations of networks adaptively', ['network structure', 'gaussian distribution'])\n", "('NONE', 'we present the XXXXX of a focused XXXXX that combines link-based and content-based approaches to predict the topical focus of an unvisited page', ['architectural design', 'web crawler'])\n", "('NONE', 'we present a custom method using the dewey decimal XXXXX to best classify the subject of an unvisited page into standard human XXXXX', ['classification system', 'knowledge categories'])\n", "('NONE', 'to prioritize an unvisited url , we use a dynamic , flexible and updating XXXXX called t-graph . it helps find the XXXXX to get to on-topic pages on the web', ['hierarchical data structure', 'shortest path'])\n", "('NONE', 'the functional and XXXXX of a focused XXXXX are elicited and described , as well as standard evaluation criteria', ['non-functional requirements', 'web crawler'])\n", "('NONE', 'the functional and XXXXX of a focused web crawler are elicited and described , as well as standard XXXXX', ['non-functional requirements', 'evaluation criteria'])\n", "('NONE', 'the functional and non-functional requirements of a focused XXXXX are elicited and described , as well as standard XXXXX', ['web crawler', 'evaluation criteria'])\n", "('NONE', 'k-anycast communication technique is utilized for enhancing transmission reliability , load-balancing and security purpose by collecting multiple copies of a packet from a source and verifying the information of monitoring field . in r-wsns , due to energy replenished continually and limited energy storage capacity , a sensor can not be always beneficial to conserve energy when a network can harvest excessive energy from the environment . therefore , the surplus energy of sensor can be used for strengthening XXXXX . in this paper , a XXXXX for k-anycast communication based upon the anycast tree scheme is proposed for wireless sensor networks', ['data transmission', 'routing protocol'])\n", "('method used for task', 'k-anycast communication technique is utilized for enhancing transmission reliability , load-balancing and security purpose by collecting multiple copies of a packet from a source and verifying the information of monitoring field . in r-wsns , due to energy replenished continually and limited energy storage capacity , a sensor can not be always beneficial to conserve energy when a network can harvest excessive energy from the environment . therefore , the surplus energy of sensor can be used for strengthening XXXXX . in this paper , a routing protocol for k-anycast communication based upon the anycast tree scheme is proposed for XXXXX', ['data transmission', 'wireless sensor networks'])\n", "('method used for task', 'k-anycast communication technique is utilized for enhancing transmission reliability , load-balancing and security purpose by collecting multiple copies of a packet from a source and verifying the information of monitoring field . in r-wsns , due to energy replenished continually and limited energy storage capacity , a sensor can not be always beneficial to conserve energy when a network can harvest excessive energy from the environment . therefore , the surplus energy of sensor can be used for strengthening data transmission . in this paper , a XXXXX for k-anycast communication based upon the anycast tree scheme is proposed for XXXXX', ['routing protocol', 'wireless sensor networks'])\n", "('method used for task', 'multiple-metrics are utilized for instructing the XXXXX . a source initiates to create a XXXXX reaching any one sink with source node as the root', ['route discovery', 'spanning tree'])\n", "('NONE', 'state the problem of XXXXX in the XXXXX', ['service composition problem', 'future iot'])\n", "('NONE', 'review the state of the art of XXXXX in the XXXXX', ['service composition', 'future iot'])\n", "('NONE', 'provide research challenges of XXXXX in the XXXXX', ['service composition', 'future iot'])\n", "('NONE', 'we address the XXXXX that arise when outsourcing XXXXX in the cloud as bpaas ( business process as a service )', ['business processes', 'security issues'])\n", "('method used for task', 'language is defined formally via XXXXX and XXXXX', ['denotational semantics', 'attribute grammars'])\n", "('NONE', 'the proposal functions of the sga in this paper are suitable to be a new standard interface because their meet the basic XXXXX of the XXXXX layer', ['security requirements', 'm2m service'])\n", "('method used for task', 'this paper investigates density , XXXXX and XXXXX numbers', ['reference point', 'position error'])\n", "('NONE', 'we present a custom method using dewey decimal XXXXX to best classify the subject of an unvisited page into standard human XXXXX', ['classification system', 'knowledge categories'])\n", "('NONE', 'to prioritize an unvisited url , we use a dynamic , flexible and updating XXXXX called t-graph , which helps find the XXXXX to get to on-topic pages on the web', ['hierarchical data structure', 'shortest path'])\n", "('NONE', 'we present a uml-based domain specific XXXXX ( dsml ) that models the sa forum availability management framework ( amf ) XXXXX', ['modeling language', 'domain model'])\n", "('NONE', 'proposal of integration of iec 61850 scl with opc ua XXXXX in a XXXXX', ['information model', 'smart grid environment'])\n", "('method used for task', 'XXXXX and XXXXX of wearable devices are addressed ,', ['product differentiation', 'product selection'])\n", "('method used for task', 'arm ( XXXXX ) is adopted to identify attractive XXXXX ,', ['association rule mining', 'design attributes'])\n", "('NONE', 'shows a mashup platform that lets XXXXX make their own XXXXX', ['end users', 'web applications'])\n", "('method used for task', 'the XXXXX was conducted for three XXXXX , ehr submissions , queries , and retrievals , based on the ihe xds.b profile for ehr sharing', ['performance testing', 'use cases'])\n", "('NONE', 'proposing the XXXXX of cdnfre tool for detecting conflicts in XXXXX', ['software architecture', 'non-functional requirements'])\n", "('method used for task', 'XXXXX to improve the XXXXX', ['expert system', 'web accessibility'])\n", "('NONE', 'XXXXX among XXXXX , regions , and images are explored', ['hierarchical structure', 'feature points'])\n", "('NONE', 'we add a XXXXX to the process of XXXXX', ['hierarchical structure', 'sparse coding'])\n", "('NONE', 'the method learns a XXXXX of geometric transformation from the XXXXX set', ['shape space', '3d image'])\n", "('method used for task', 'XXXXX is registration between planning time and XXXXX in radiotherapy', ['target problem', 'treatment time'])\n", "('NONE', 'registration is performed on XXXXX of 17 patients with XXXXX', ['mr images', 'cervical cancer'])\n", "('NONE', 'XXXXX is developed in a variational framework using XXXXX', ['segmentation algorithm', 'level sets'])\n", "('NONE', 'this framework is based on XXXXX of 2d closed , XXXXX', ['shape analysis', 'planar curves'])\n", "('method used for task', 'the models can be used in XXXXX and 3d XXXXX', ['image registration', 'surface reconstruction'])\n", "('method used for task', 'the method is not overly sensitive with respect to XXXXX , missing frames and XXXXX', ['missing data', 'measurement noise'])\n", "('NONE', 'incorporate the XXXXX into an optimized search based XXXXX', ['shape model', 'segmentation method'])\n", "('method used for task', 'a XXXXX based on XXXXX is proposed', ['video segmentation', 'dynamic texture'])\n", "('method used for task', 'comparing with XXXXX , higher chance to find the XXXXX is achieved', ['chan–vese model', 'global solution'])\n", "('method used for task', 'we incorporate the spatial context to the conventional XXXXX ( fcm ) algorithm for XXXXX', ['fuzzy c-means', 'image segmentation'])\n", "('NONE', 'we propose a method for XXXXX from single XXXXX', ['pose estimation', 'ultrasound image'])\n", "('method used for task', 'a XXXXX correspondence algorithm is proposed for visual–thermal close-range XXXXX', ['video surveillance', 'stereo dense'])\n", "('NONE', 'such elastic XXXXX serve as XXXXX in a bayesian active contour model', ['shape models', 'shape priors'])\n", "('NONE', 'such elastic XXXXX serve as shape priors in a bayesian XXXXX', ['shape models', 'active contour model'])\n", "('NONE', 'such elastic shape models serve as XXXXX in a bayesian XXXXX', ['shape priors', 'active contour model'])\n", "('method used for task', 'we use a global XXXXX to segment leaves in complex XXXXX', ['shape model', 'natural images'])\n", "('method used for task', 'XXXXX and XXXXX improve segmentation performance', ['prior knowledge', 'shape constraints'])\n", "('method used for task', 'XXXXX and shape constraints improve XXXXX', ['prior knowledge', 'segmentation performance'])\n", "('NONE', 'prior knowledge and XXXXX improve XXXXX', ['shape constraints', 'segmentation performance'])\n", "('NONE', 'we present a comprehensive survey of XXXXX ( mrfs ) in XXXXX', ['markov random fields', 'computer vision'])\n", "('NONE', 'local memorization guides to search for the XXXXX that avoids XXXXX', ['optimal solution', 'local minima'])\n", "('method used for task', 'we describe and illustrate an XXXXX , based on the developed aggm , for XXXXX using infrared images', ['online algorithm', 'pedestrian detection'])\n", "('NONE', 'pca is the best XXXXX in very XXXXX less than 16*16pixels', ['recognition rate', 'low resolution'])\n", "('method used for task', 'it fits ellipses to XXXXX to analyse full XXXXX and motion', ['body parts', 'body shape'])\n", "('method used for task', 'opn could be utilized as XXXXX with semantics for scalable XXXXX', ['visual words', 'image retrieval'])\n", "('method used for task', 'extract XXXXX by an effective paradigm for XXXXX', ['semantic information', 'saliency detection'])\n", "('method used for task', 'a XXXXX based manifold representation for XXXXX', ['local feature', 'object tracking'])\n", "('method used for task', 'experimental results on both the synthetic XXXXX and recent ir database demonstrate jvim’s advantages over several recent XXXXX', ['shape models', 'cad data'])\n", "('method used for task', 'the data is calculated using gis-based XXXXX and XXXXX', ['augmented reality', 'computer vision'])\n", "('NONE', 'we use random walks to perform XXXXX with XXXXX', ['image segmentation', 'directed hypergraphs'])\n", "('method used for task', 'we propose a XXXXX for solving the XXXXX model', ['hybrid method', 'mumford–shah segmentation'])\n", "('method used for task', 'we show that the XXXXX is efficient for a wide range of XXXXX', ['hybrid method', 'model parameters'])\n", "('NONE', 'our results are close to XXXXX in XXXXX', ['human performance', 'expression recognition'])\n", "('method used for task', 'we applied multidimensional XXXXX , the matching metric , and a XXXXX to classify hepatic lesions', ['persistent homology', 'support vector machine'])\n", "('method used for task', 'application of XXXXX to the evaluation of a method for XXXXX', ['persistent homology', '3d reconstruction'])\n", "('NONE', 'reduce the length of the XXXXX using XXXXX', ['time series', 'dimensionality reduction techniques'])\n", "('method used for task', 'XXXXX for XXXXX', ['unsupervised learning', 'feature selection'])\n", "('NONE', 'our masks are based on XXXXX using a linear XXXXX', ['feature selection', 'support vector machine'])\n", "('method used for task', 'a XXXXX is linearized by rank relaxation to reduce the XXXXX', ['bilinear model', 'time complexity'])\n", "('method used for task', 'a novel XXXXX for the XXXXX is incorporated', ['uncertainty measure', 'motion estimation'])\n", "('method used for task', 'presents novel neuroscience inspired information theoretic approach to XXXXX based on XXXXX', ['motion segmentation', 'mutual information'])\n", "('NONE', 'comparative XXXXX against competing XXXXX', ['performance evaluation', 'segmentation methods'])\n", "('method used for task', 'XXXXX on the XXXXX are derived from affine correspondences', ['linear constraints', 'fundamental matrix'])\n", "('NONE', 'a XXXXX of the XXXXX by intersection of conics is proposed', ['fundamental matrix', 'calculation method'])\n", "('NONE', 'glss is built on global and local XXXXX that help boosting the efficacy of XXXXX', ['data structures', 'feature selection'])\n", "('method used for task', 'fast knn XXXXX is well integrated with submodular XXXXX', ['graph construction', 'dictionary learning'])\n", "('NONE', 'we present an objective XXXXX of statistical XXXXX', ['performance analysis', 'edge detection'])\n", "('NONE', 'we show how XXXXX can outperform traditional XXXXX', ['statistical tests', 'edge detection methods'])\n", "('NONE', 'introduce XXXXX into the XXXXX of quantization', ['statistical analysis', 'out-of-sample extension'])\n", "('method used for task', 'a semi-supervised multi-graph XXXXX is proposed for XXXXX', ['image search', 'hashing method'])\n", "('NONE', 'rao-blackwellized XXXXX with XXXXX', ['particle filtering', 'gaussian mixture models'])\n", "('NONE', 'the core contribution is the XXXXX of the XXXXX', ['formal proof', 'bijection principle'])\n", "('method used for task', 'two XXXXX , one on manifolds for appearance learning , another for XXXXX', ['particle filters', 'object tracking'])\n", "('method used for task', 'XXXXX to prevent XXXXX when objects are likely occluded', ['occlusion handling', 'online learning'])\n", "('NONE', 'our XXXXX operates directly on XXXXX via the sampson distance', ['cost function', 'image points'])\n", "('NONE', 'it is validated in XXXXX , XXXXX and shape retrieval scenarios', ['object recognition', '3d reconstruction'])\n", "('NONE', 'XXXXX of two distinctive examplars of XXXXX', ['theoretical analysis', 'statistical shape models'])\n", "('NONE', 'extensive XXXXX of two distinctive examplars of XXXXX', ['experimental comparison', 'statistical shape models'])\n", "('method used for task', 'we propose a new slice representation model of XXXXX for XXXXX', ['range data', 'head pose estimation'])\n", "('method used for task', 'we present a XXXXX to XXXXX for freely moving cameras', ['geometric approach', 'background subtraction'])\n", "('NONE', 'XXXXX are detected in complex scenes with significant XXXXX', ['moving objects', 'camera motion'])\n", "('method used for task', 'to do so we learn 3d XXXXX between objects and different XXXXX', ['spatial relationships', 'body parts'])\n", "('method used for task', 'we developed new XXXXX based on the XXXXX , hts and htsn', ['shape descriptors', 'hough transform'])\n", "('method used for task', 'a new XXXXX for XXXXX is proposed', ['evaluation procedure', 'action localization'])\n", "('NONE', 'a single XXXXX integrates out XXXXX', ['performance measure', 'quality constraints'])\n", "('NONE', 'soft XXXXX estimated from XXXXX', ['upper bounds', 'experimental data'])\n", "('NONE', 'a novel approach to determine adaptively the XXXXX of XXXXX', ['temporal consistency', 'particle filters'])\n", "('method used for task', 'proposed method provides satisfactory XXXXX and XXXXX', ['detection accuracy', 'generalization capability'])\n", "('method used for task', 'indepth XXXXX and XXXXX', ['experimental validation', 'error analysis'])\n", "('NONE', 'dominant XXXXX by exploring XXXXX with histogram', ['gradient information', 'pixel selection'])\n", "('NONE', 'we encode pairwise relative XXXXX of patches in the bag of XXXXX', ['spatial information', 'word model'])\n", "('NONE', 'a new XXXXX optimized with XXXXX is proposed', ['objective function', 'graph cuts'])\n", "('method used for task', 'new and well-known XXXXX and XXXXX are both used', ['feature extraction', 'classification methods'])\n", "('method used for task', 'object matching and XXXXX is combined for achieving automatic XXXXX', ['active contour', 'object segmentation'])\n", "('method used for task', 'XXXXX based on likelihood and XXXXX without line search', ['learning algorithms', 'margin maximization'])\n", "('NONE', 'we formulate XXXXX as a statistical XXXXX', ['image segmentation', 'parameter estimation problem'])\n", "('method used for task', 'thorough XXXXX to compare the performances of several algorithms for hierarchical XXXXX', ['experimental study', 'image classification'])\n", "('NONE', 'we recover simultaneously the non-rigid XXXXX and the corresponding camera pose from a single image in XXXXX', ['3d shape', 'real time'])\n", "('NONE', 'high XXXXX in two computer vision applications : XXXXX and human action recognition', ['classification accuracy', 'facial expression recognition'])\n", "('NONE', 'high XXXXX in two computer vision applications : facial expression recognition and XXXXX', ['classification accuracy', 'human action recognition'])\n", "('method used for task', 'high classification accuracy in two computer vision applications : XXXXX and XXXXX', ['facial expression recognition', 'human action recognition'])\n", "('NONE', 'projection of the XXXXX from the XXXXX to the tangent euclidean space', ['covariance matrix', 'riemannian manifold'])\n", "('method used for task', 'projection of the XXXXX from the riemannian manifold to the tangent XXXXX', ['covariance matrix', 'euclidean space'])\n", "('method used for task', 'projection of the covariance matrix from the XXXXX to the tangent XXXXX', ['riemannian manifold', 'euclidean space'])\n", "('NONE', 'the algorithm is efficient both in terms of XXXXX and of XXXXX', ['computational cost', 'detection performance'])\n", "('NONE', 'based on XXXXX , a hybrid XXXXX is proposed', ['kernel regression', 'fusion strategy'])\n", "('method used for task', 'a XXXXX for each XXXXX', ['closed-form solution', 'camera model'])\n", "('method used for task', 'an XXXXX comparing the algebraic procedures to global polynomial optimization and an XXXXX', ['experimental evaluation', 'interior-point method'])\n", "('NONE', 'this method leverages discriminative computer vision models for faster XXXXX in XXXXX', ['probabilistic inference', 'generative models'])\n", "('NONE', 'a two-stage XXXXX system using XXXXX is proposed', ['head detection', 'motion features'])\n", "('method used for task', 'geometric normalization is critical to most XXXXX ( fr ) algorithms and is usually based off XXXXX', ['face recognition', 'eye locations'])\n", "('NONE', 'the XXXXX obtained from the proposed algorithm out perform other tested algorithms in both XXXXX and fr results', ['eye detection', 'eye locations'])\n", "('method used for task', 'a new 3d local XXXXX for XXXXX and human action recognition from depth sensors', ['shape descriptor', 'hand gesture'])\n", "('method used for task', 'a new 3d local XXXXX for hand gesture and XXXXX from depth sensors', ['shape descriptor', 'human action recognition'])\n", "('method used for task', 'a new 3d local shape descriptor for XXXXX and XXXXX from depth sensors', ['hand gesture', 'human action recognition'])\n", "('method used for task', 'XXXXX for the direct XXXXX based camera calibration', ['uncertainty analysis', 'linear transformation'])\n", "('method used for task', 'XXXXX for the direct linear transformation based XXXXX', ['uncertainty analysis', 'camera calibration'])\n", "('NONE', 'uncertainty analysis for the direct XXXXX based XXXXX', ['linear transformation', 'camera calibration'])\n", "('method used for task', 'we propose a method for XXXXX on 3d and 4d XXXXX', ['range data', 'landmark localization'])\n", "('NONE', 'we study object motion in stereo XXXXX by providing a novel XXXXX', ['mathematical analysis', 'video content'])\n", "('method used for task', 'the XXXXX is described by weighted XXXXX', ['object model', 'gaussian mixture models'])\n", "('NONE', 'sin parses structured XXXXX ( or XXXXX in general )', ['time series data', 'activity sequence'])\n", "('method used for task', 'hand-gesture XXXXX based on XXXXX for hci', ['recognition system', 'color imagery'])\n", "('NONE', 'a redundant XXXXX has been composed by using XXXXX', ['wavelet transform', 'texture feature space'])\n", "('NONE', 'the XXXXX method resembles dense surface shape recovery from XXXXX', ['missing data', '3d face estimation'])\n", "('method used for task', 'combining XXXXX and XXXXX with multiplication performs better for nearly planar objects', ['edge information', 'surface normals'])\n", "('method used for task', 'two well-known state-of-the-art methods and one XXXXX are extended to use this XXXXX', ['semantic information', 'baseline method'])\n", "('NONE', 'novel aam based XXXXX and combination of XXXXX and XXXXX', ['hand gesture', 'face features'])\n", "('NONE', 'novel aam based XXXXX and combination of XXXXX and XXXXX', ['hand gesture', 'face features'])\n", "('method used for task', 'we propose a hierarchical kernel based method to learn the non-linear correlations between rgb and depth XXXXX for XXXXX', ['action recognition', 'action data'])\n", "('method used for task', 'part models are formulated with a XXXXX for XXXXX', ['graph structure', 'object tracking'])\n", "('NONE', 'XXXXX are formulated with a XXXXX for object tracking', ['graph structure', 'part models'])\n", "('method used for task', 'XXXXX are formulated with a graph structure for XXXXX', ['object tracking', 'part models'])\n", "('method used for task', 'weighted XXXXX are used to handle XXXXX change and occlusion', ['part models', 'target appearance'])\n", "('method used for task', 'XXXXX are used to control the sample selections for XXXXX update', ['weight models', 'part model'])\n", "('method used for task', 'we experiment on a high variety of scenarios and public datasets ( genre classification - blip10000 , XXXXX - ucf50 / ucf101 and daily XXXXX - adl ) and show the benefits of the proposed approach which outperforms other state of the art approaches', ['action recognition', 'activities recognition'])\n", "('NONE', 'coupling detection and XXXXX of XXXXX in a single function', ['data association', 'object tracking'])\n", "('method used for task', 'XXXXX is incorporated with XXXXX by extracting the interactive contours', ['contextual information', 'motion features'])\n", "('NONE', 'a latent XXXXX integrating XXXXX , group discovery , and activity recognition is proposed', ['graphical model', 'multi-target tracking'])\n", "('method used for task', 'a latent XXXXX integrating multi-target tracking , group discovery , and XXXXX is proposed', ['graphical model', 'activity recognition'])\n", "('method used for task', 'a latent graphical model integrating XXXXX , group discovery , and XXXXX is proposed', ['multi-target tracking', 'activity recognition'])\n", "('NONE', 'performance of XXXXX improves when XXXXX and group clustering are incorporated', ['activity recognition', 'multi-target tracking'])\n", "('method used for task', 'our method showed superior performance on both XXXXX and realistic dataset on action and XXXXX', ['synthetic data', 'gesture recognition'])\n", "('NONE', 'the framework allows the creation of a high-level XXXXX of the scene and scalable XXXXX', ['semantic representation', 'feature extraction'])\n", "('method used for task', 'an XXXXX for inference in binary XXXXX mrf–map is proposed', ['efficient algorithm', 'higher order'])\n", "('method used for task', 'a XXXXX is formulated as a XXXXX discovering problem', ['multi-target tracking', 'dense subgraph'])\n", "('NONE', 'XXXXX : XXXXX are clearly visible in different light conditions', ['robust solution', 'traffic lights'])\n", "('method used for task', 'XXXXX can provide semantic cue to enhance the XXXXX', ['object recognition', 'material recognition'])\n", "('NONE', 'dense subgraphs convey more information about local XXXXX than simple XXXXX', ['graph structure', 'centrality measures'])\n", "('NONE', 'XXXXX convey more information about local XXXXX than simple centrality measures', ['graph structure', 'dense subgraphs'])\n", "('NONE', 'XXXXX convey more information about local graph structure than simple XXXXX', ['centrality measures', 'dense subgraphs'])\n", "('method used for task', 'region compactness in addition to XXXXX and XXXXX has been used to compute image region similarities', ['color features', 'image intensity'])\n", "('method used for task', 'the framework yields a simple , efficient , XXXXX for XXXXX', ['closed-form solution', 'change detection'])\n", "('method used for task', 'we detach the intrinsic XXXXX related to either brightness or XXXXX', ['camera parameters', 'depth data'])\n", "('NONE', 'the adce algorithm can find the salient XXXXX of XXXXX', ['feature points', 'shape contour'])\n", "('method used for task', 'the paper proposes a geometric error based triangulation and metric study of lines . as one of the fundamental problems in XXXXX , line triangulation is to determine the 3d coordinates of a line based on its 2d image projections from more than two views of cameras . compared to point features , XXXXX are more robust to matching errors , occlusions , and image uncertainties . however , when the number of views is larger than two , the back-projection planes usually do not intersect at one line due to measurement errors and image noise . thus , it is critical to find a 3d line that optimally fits the measured data . this paper , at first time , provides a comprehensive study on the triangulation and metric of lines based on geometric error', ['computer vision', 'line segments'])\n", "('method used for task', 'the paper proposes a geometric error based triangulation and metric study of lines . as one of the fundamental problems in XXXXX , line triangulation is to determine the 3d coordinates of a line based on its 2d image projections from more than two views of cameras . compared to point features , line segments are more robust to matching errors , occlusions , and image uncertainties . however , when the number of views is larger than two , the back-projection planes usually do not intersect at one line due to XXXXX and image noise . thus , it is critical to find a 3d line that optimally fits the measured data . this paper , at first time , provides a comprehensive study on the triangulation and metric of lines based on geometric error', ['computer vision', 'measurement errors'])\n", "('NONE', 'the paper proposes a geometric error based triangulation and metric study of lines . as one of the fundamental problems in XXXXX , XXXXX is to determine the 3d coordinates of a line based on its 2d image projections from more than two views of cameras . compared to point features , line segments are more robust to matching errors , occlusions , and image uncertainties . however , when the number of views is larger than two , the back-projection planes usually do not intersect at one line due to measurement errors and image noise . thus , it is critical to find a 3d line that optimally fits the measured data . this paper , at first time , provides a comprehensive study on the triangulation and metric of lines based on geometric error', ['computer vision', 'line triangulation'])\n", "('method used for task', 'the paper proposes a geometric error based triangulation and metric study of lines . as one of the fundamental problems in computer vision , line triangulation is to determine the 3d coordinates of a line based on its 2d image projections from more than two views of cameras . compared to point features , XXXXX are more robust to matching errors , occlusions , and image uncertainties . however , when the number of views is larger than two , the back-projection planes usually do not intersect at one line due to XXXXX and image noise . thus , it is critical to find a 3d line that optimally fits the measured data . this paper , at first time , provides a comprehensive study on the triangulation and metric of lines based on geometric error', ['line segments', 'measurement errors'])\n", "('method used for task', 'the paper proposes a geometric error based triangulation and metric study of lines . as one of the fundamental problems in computer vision , XXXXX is to determine the 3d coordinates of a line based on its 2d image projections from more than two views of cameras . compared to point features , XXXXX are more robust to matching errors , occlusions , and image uncertainties . however , when the number of views is larger than two , the back-projection planes usually do not intersect at one line due to measurement errors and image noise . thus , it is critical to find a 3d line that optimally fits the measured data . this paper , at first time , provides a comprehensive study on the triangulation and metric of lines based on geometric error', ['line segments', 'line triangulation'])\n", "('method used for task', 'the paper proposes a geometric error based triangulation and metric study of lines . as one of the fundamental problems in computer vision , XXXXX is to determine the 3d coordinates of a line based on its 2d image projections from more than two views of cameras . compared to point features , line segments are more robust to matching errors , occlusions , and image uncertainties . however , when the number of views is larger than two , the back-projection planes usually do not intersect at one line due to XXXXX and image noise . thus , it is critical to find a 3d line that optimally fits the measured data . this paper , at first time , provides a comprehensive study on the triangulation and metric of lines based on geometric error', ['measurement errors', 'line triangulation'])\n", "('method used for task', 'in this paper , a comprehensive study of line triangulation is conducted using geometric XXXXX . compared to the algebraic error based approaches , geometric error based algorithm is more meaningful , and thus , yields better estimation results . the main contributions of this study include : ( i ) it is proved that the XXXXX to minimizing the geometric errors can be transformed to finding the real roots of algebraic equations ; ( ii ) an effective iterative algorithm , iteg , is proposed to minimizing the geometric errors ; and ( iii ) an in-depth comparative evaluations on three metrics in 3d line space , the euclidean metric , the orthogonal metric , and the quasi-riemannian metric , are carried out . extensive experiments on synthetic data and real images are carried out to validate and demonstrate the effectiveness of the proposed algorithms', ['cost functions', 'optimal solution'])\n", "('method used for task', 'in this paper , a comprehensive study of line triangulation is conducted using geometric XXXXX . compared to the algebraic error based approaches , geometric error based algorithm is more meaningful , and thus , yields better estimation results . the main contributions of this study include : ( i ) it is proved that the optimal solution to minimizing the geometric errors can be transformed to finding the real roots of algebraic equations ; ( ii ) an effective XXXXX , iteg , is proposed to minimizing the geometric errors ; and ( iii ) an in-depth comparative evaluations on three metrics in 3d line space , the euclidean metric , the orthogonal metric , and the quasi-riemannian metric , are carried out . extensive experiments on synthetic data and real images are carried out to validate and demonstrate the effectiveness of the proposed algorithms', ['cost functions', 'iterative algorithm'])\n", "('method used for task', 'in this paper , a comprehensive study of line triangulation is conducted using geometric XXXXX . compared to the algebraic error based approaches , geometric error based algorithm is more meaningful , and thus , yields better estimation results . the main contributions of this study include : ( i ) it is proved that the optimal solution to minimizing the geometric errors can be transformed to finding the real roots of algebraic equations ; ( ii ) an effective iterative algorithm , iteg , is proposed to minimizing the geometric errors ; and ( iii ) an in-depth comparative evaluations on three metrics in 3d line space , the euclidean metric , the orthogonal metric , and the quasi-riemannian metric , are carried out . extensive experiments on XXXXX and real images are carried out to validate and demonstrate the effectiveness of the proposed algorithms', ['cost functions', 'synthetic data'])\n", "('method used for task', 'in this paper , a comprehensive study of XXXXX is conducted using geometric XXXXX . compared to the algebraic error based approaches , geometric error based algorithm is more meaningful , and thus , yields better estimation results . the main contributions of this study include : ( i ) it is proved that the optimal solution to minimizing the geometric errors can be transformed to finding the real roots of algebraic equations ; ( ii ) an effective iterative algorithm , iteg , is proposed to minimizing the geometric errors ; and ( iii ) an in-depth comparative evaluations on three metrics in 3d line space , the euclidean metric , the orthogonal metric , and the quasi-riemannian metric , are carried out . extensive experiments on synthetic data and real images are carried out to validate and demonstrate the effectiveness of the proposed algorithms', ['cost functions', 'line triangulation'])\n", "('method used for task', 'in this paper , a comprehensive study of line triangulation is conducted using geometric cost functions . compared to the algebraic error based approaches , geometric error based algorithm is more meaningful , and thus , yields better estimation results . the main contributions of this study include : ( i ) it is proved that the XXXXX to minimizing the geometric errors can be transformed to finding the real roots of algebraic equations ; ( ii ) an effective XXXXX , iteg , is proposed to minimizing the geometric errors ; and ( iii ) an in-depth comparative evaluations on three metrics in 3d line space , the euclidean metric , the orthogonal metric , and the quasi-riemannian metric , are carried out . extensive experiments on synthetic data and real images are carried out to validate and demonstrate the effectiveness of the proposed algorithms', ['optimal solution', 'iterative algorithm'])\n", "('method used for task', 'in this paper , a comprehensive study of line triangulation is conducted using geometric cost functions . compared to the algebraic error based approaches , geometric error based algorithm is more meaningful , and thus , yields better estimation results . the main contributions of this study include : ( i ) it is proved that the XXXXX to minimizing the geometric errors can be transformed to finding the real roots of algebraic equations ; ( ii ) an effective iterative algorithm , iteg , is proposed to minimizing the geometric errors ; and ( iii ) an in-depth comparative evaluations on three metrics in 3d line space , the euclidean metric , the orthogonal metric , and the quasi-riemannian metric , are carried out . extensive experiments on XXXXX and real images are carried out to validate and demonstrate the effectiveness of the proposed algorithms', ['optimal solution', 'synthetic data'])\n", "('method used for task', 'in this paper , a comprehensive study of XXXXX is conducted using geometric cost functions . compared to the algebraic error based approaches , geometric error based algorithm is more meaningful , and thus , yields better estimation results . the main contributions of this study include : ( i ) it is proved that the XXXXX to minimizing the geometric errors can be transformed to finding the real roots of algebraic equations ; ( ii ) an effective iterative algorithm , iteg , is proposed to minimizing the geometric errors ; and ( iii ) an in-depth comparative evaluations on three metrics in 3d line space , the euclidean metric , the orthogonal metric , and the quasi-riemannian metric , are carried out . extensive experiments on synthetic data and real images are carried out to validate and demonstrate the effectiveness of the proposed algorithms', ['optimal solution', 'line triangulation'])\n", "('method used for task', 'in this paper , a comprehensive study of line triangulation is conducted using geometric cost functions . compared to the algebraic error based approaches , geometric error based algorithm is more meaningful , and thus , yields better estimation results . the main contributions of this study include : ( i ) it is proved that the optimal solution to minimizing the geometric errors can be transformed to finding the real roots of algebraic equations ; ( ii ) an effective XXXXX , iteg , is proposed to minimizing the geometric errors ; and ( iii ) an in-depth comparative evaluations on three metrics in 3d line space , the euclidean metric , the orthogonal metric , and the quasi-riemannian metric , are carried out . extensive experiments on XXXXX and real images are carried out to validate and demonstrate the effectiveness of the proposed algorithms', ['iterative algorithm', 'synthetic data'])\n", "('method used for task', 'in this paper , a comprehensive study of XXXXX is conducted using geometric cost functions . compared to the algebraic error based approaches , geometric error based algorithm is more meaningful , and thus , yields better estimation results . the main contributions of this study include : ( i ) it is proved that the optimal solution to minimizing the geometric errors can be transformed to finding the real roots of algebraic equations ; ( ii ) an effective XXXXX , iteg , is proposed to minimizing the geometric errors ; and ( iii ) an in-depth comparative evaluations on three metrics in 3d line space , the euclidean metric , the orthogonal metric , and the quasi-riemannian metric , are carried out . extensive experiments on synthetic data and real images are carried out to validate and demonstrate the effectiveness of the proposed algorithms', ['iterative algorithm', 'line triangulation'])\n", "('method used for task', 'in this paper , a comprehensive study of XXXXX is conducted using geometric cost functions . compared to the algebraic error based approaches , geometric error based algorithm is more meaningful , and thus , yields better estimation results . the main contributions of this study include : ( i ) it is proved that the optimal solution to minimizing the geometric errors can be transformed to finding the real roots of algebraic equations ; ( ii ) an effective iterative algorithm , iteg , is proposed to minimizing the geometric errors ; and ( iii ) an in-depth comparative evaluations on three metrics in 3d line space , the euclidean metric , the orthogonal metric , and the quasi-riemannian metric , are carried out . extensive experiments on XXXXX and real images are carried out to validate and demonstrate the effectiveness of the proposed algorithms', ['synthetic data', 'line triangulation'])\n", "('NONE', 'a system with dynamic lighting is proposed that predicts the validity of a photometric model without XXXXX of the XXXXX', ['prior knowledge', 'scene geometry'])\n", "('method used for task', 'a XXXXX to improve the XXXXX given the user’s position , available sensors , and recognition tools is presented', ['speech recognition', 'fusion method'])\n", "('NONE', 'gui for clinical uses allows real-time , XXXXX with minimal XXXXX', ['high performance', 'user feedback'])\n", "('method used for task', 'XXXXX are adopted for tooth brushing XXXXX', ['hidden markov models', 'gesture recognition'])\n", "('NONE', 'we develop a XXXXX as a type of XXXXX', ['assistive technology', 'hci system'])\n", "('method used for task', 'we introduce a new dataset , the video water database , for XXXXX and to encourage research into XXXXX', ['experimental evaluation', 'water detection'])\n", "('method used for task', 'we show experimentally that our water detection method improves over methods from XXXXX and XXXXX', ['dynamic texture', 'material recognition'])\n", "('NONE', 'we show experimentally that our XXXXX method improves over methods from XXXXX and material recognition', ['dynamic texture', 'water detection'])\n", "('method used for task', 'we show experimentally that our XXXXX method improves over methods from dynamic texture and XXXXX', ['material recognition', 'water detection'])\n", "('method used for task', 'online XXXXX and quality visualization for XXXXX', ['user feedback', 'image acquisition'])\n", "('method used for task', 'we revisit three tasks : image sensing , XXXXX and XXXXX', ['scene segmentation', 'optical flow'])\n", "('NONE', 'a methodology for detecting the XXXXX position in XXXXX is presented', ['fundus images', 'fovea center'])\n", "('method used for task', 'XXXXX : methodology-provided and actual XXXXX distance', ['evaluation criterion', 'fovea center'])\n", "('method used for task', 'advanced texture analysis methods – XXXXX and gaussian XXXXX', ['local binary patterns', 'markov random fields'])\n", "('NONE', 'early XXXXX with lsa outperforms late XXXXX', ['data fusion', 'fusion methods'])\n", "('NONE', 'low-level fusion of the XXXXX slightly improves the XXXXX', ['visual features', 'predictive performance'])\n", "('method used for task', 'as a result of improved XXXXX compared to previous methods , needle XXXXX and robustness are increased using the proposed method', ['segmentation performance', 'localization accuracy'])\n", "('method used for task', 'the area under XXXXX increased from 0.78 to 0.86 with the XXXXX', ['roc curve', 'data-driven approach'])\n", "('NONE', 'XXXXX that significantly reduces the XXXXX', ['efficient implementation', 'processing time'])\n", "('method used for task', 'a new two-layer structural XXXXX for detecting microscopic image cells , which focuses on the representation issue to propose a model to effectively capture the rich XXXXX', ['contextual information', 'prediction framework'])\n", "('method used for task', 'we presented an XXXXX that is tailored to model and analyze the two-channel images of XXXXX', ['image processing pipeline', 'muscle fibers'])\n", "('NONE', 'the method quantifies morphological and XXXXX of nuclei and cytoplasm of XXXXX and derives other measurements about XXXXX', ['geometric features', 'muscle fibers'])\n", "('NONE', 'the method quantifies morphological and XXXXX of nuclei and cytoplasm of XXXXX and derives other measurements about XXXXX', ['geometric features', 'muscle fibers'])\n", "('method used for task', 'a detailed review of automated XXXXX for the pulmonary lobes from clinical XXXXX', ['segmentation methods', 'ct data'])\n", "('method used for task', 'we proposed a XXXXX based method to detect the respiratory signal from 2d XXXXX', ['manifold learning', 'ultrasound images'])\n", "('NONE', 'we propose the voxel visibility model for XXXXX in XXXXX', ['transfer function', 'volume rendering'])\n", "('method used for task', 'we propose the XXXXX model for XXXXX in volume rendering', ['transfer function', 'voxel visibility'])\n", "('method used for task', 'we propose the XXXXX model for transfer function in XXXXX', ['volume rendering', 'voxel visibility'])\n", "('method used for task', 'we propose an XXXXX for automatic XXXXX', ['optimization algorithm', 'transfer function design'])\n", "('method used for task', 'memory based XXXXX that is robust to pixel-level XXXXX', ['active contour', 'classification errors'])\n", "('NONE', 'test 7180 crypts : 87 % and 9 % true and XXXXX , 96 % XXXXX', ['false positives', 'segmentation accuracy'])\n", "('method used for task', 'evaluations over a rv XXXXX and hlhs XXXXX show the robustness of the method', ['challenge data', 'patient data'])\n", "('NONE', 'we perform XXXXX of healthy and tumor areas in XXXXX of bcc skin samples', ['automated classification', 'cars images'])\n", "('NONE', 'we believe this is an important step towards automated XXXXX in XXXXX', ['tumor detection', 'cars images'])\n", "('NONE', 'implementation of XXXXX in a XXXXX organ extraction procedure', ['granular computing', 'model based'])\n", "('method used for task', 'a XXXXX is used for XXXXX and svm parameter selection', ['genetic algorithm', 'feature selection'])\n", "('NONE', 'XXXXX in XXXXX tasks', ['experimental validation', 'cocaine addiction'])\n", "('method used for task', 'building an XXXXX able to process the XXXXX', ['expert system', 'feature vector'])\n", "('method used for task', 'XXXXX and XXXXX are used during feature extraction', ['principal component analysis', 'radon transform'])\n", "('method used for task', 'XXXXX and radon transform are used during XXXXX', ['principal component analysis', 'feature extraction'])\n", "('method used for task', 'principal component analysis and XXXXX are used during XXXXX', ['radon transform', 'feature extraction'])\n", "('NONE', 'we analyzed the XXXXX after XXXXX', ['healing process', 'bone loss'])\n", "('NONE', 'XXXXX decreases during the XXXXX after bone loss', ['fractal dimension', 'healing process'])\n", "('NONE', 'XXXXX decreases during the healing process after XXXXX', ['fractal dimension', 'bone loss'])\n", "('NONE', 'fractal dimension decreases during the XXXXX after XXXXX', ['healing process', 'bone loss'])\n", "('method used for task', 'a fully automatic XXXXX is presented to detect XXXXX using mri', ['cad system', 'prostate cancer'])\n", "('NONE', 'XXXXX have an advantage over XXXXX', ['mr images', 'ct images'])\n", "('NONE', 'we improved XXXXX by adaptation of additional training data and interpolating XXXXX', ['language models', 'translation quality'])\n", "('method used for task', 'system aims to provide longitudinal tracking , XXXXX , and XXXXX', ['data mining', 'data analysis'])\n", "('NONE', 'choroidal thickness and XXXXX are given vis-à-vis XXXXX', ['volume estimates', 'observer repeatability'])\n", "('method used for task', 'XXXXX had little dependency on the time after XXXXX', ['quantitative imaging features', 'contrast injection'])\n", "('NONE', 'some XXXXX changed gradually after XXXXX but eventually became stable', ['texture features', 'contrast injection'])\n", "('method used for task', 'energy functional includes XXXXX for XXXXX and level set function', ['total variation regularization', 'bias field'])\n", "('method used for task', 'energy functional includes XXXXX for bias field and XXXXX', ['total variation regularization', 'level set function'])\n", "('method used for task', 'energy functional includes total variation regularization for XXXXX and XXXXX', ['bias field', 'level set function'])\n", "('method used for task', 'a multiscale XXXXX for XXXXX is proposed', ['region growing method', 'coronary artery segmentation'])\n", "('method used for task', 'the first multi-center milestone XXXXX for vertebra XXXXX', ['comparative study', 'segmentation methods'])\n", "('method used for task', 'a new energy aware XXXXX has been proposed for cluster based XXXXX', ['routing algorithm', 'wireless sensor networks'])\n", "('NONE', 'it achieves o ( 1 ) XXXXX per XXXXX and o ( n ) time complexity for a wsn having n sensor nodes', ['message complexity', 'sensor node'])\n", "('method used for task', 'it achieves o ( 1 ) XXXXX per sensor node and o ( n ) XXXXX for a wsn having n sensor nodes', ['message complexity', 'time complexity'])\n", "('method used for task', 'it achieves o ( 1 ) XXXXX per sensor node and o ( n ) time complexity for a wsn having n XXXXX', ['message complexity', 'sensor nodes'])\n", "('method used for task', 'it achieves o ( 1 ) message complexity per XXXXX and o ( n ) XXXXX for a wsn having n sensor nodes', ['sensor node', 'time complexity'])\n", "('method used for task', 'it achieves o ( 1 ) message complexity per XXXXX and o ( n ) time complexity for a wsn having n XXXXX', ['sensor node', 'sensor nodes'])\n", "('method used for task', 'it achieves o ( 1 ) message complexity per sensor node and o ( n ) XXXXX for a wsn having n XXXXX', ['time complexity', 'sensor nodes'])\n", "('method used for task', 'it is successful in balancing the relaying load among the XXXXX with respect to their XXXXX', ['residual energy', 'sensor nodes'])\n", "('method used for task', 'XXXXX is solved by XXXXX and compared the performance with pso and fa', ['differential evolution algorithm', 'sa problem'])\n", "('method used for task', 'face allows XXXXX to customize XXXXX , localization , and processing procedures', ['data partitioning', 'application developers'])\n", "('method used for task', 'establishing the security-aware task , security overhead and XXXXX for aperiodic XXXXX', ['real-time applications', 'risk models'])\n", "('NONE', 'we achieve the XXXXX without inviting large XXXXX', ['storage capacity', 'communication overhead'])\n", "('method used for task', 'an sdma-based XXXXX ( s-mac ) for XXXXX is proposed', ['mac protocol', 'wireless ad hoc networks'])\n", "('method used for task', 'the proposed method is novel as it is important to install low cost sensors and XXXXX along with induction machines to achieve short XXXXX and an automated way of reporting the fault', ['detection mechanisms', 'detection time'])\n", "('NONE', 'a novel XXXXX 12t mtcmos based XXXXX is proposed', ['low power', 'sram cell'])\n", "('method used for task', 'XXXXX reduces the effect on XXXXX caused by illumination', ['histogram equalization', 'msa features'])\n", "('method used for task', 'the number of ( α , β ) pairs affects the size of XXXXX and XXXXX', ['msa features', 'recognition rates'])\n", "('method used for task', 'we propose semantic-based XXXXX through deductive logic and XXXXX', ['policy analysis', 'inference rules'])\n", "('method used for task', 'we present flaw , conflict and XXXXX algorithms for XXXXX', ['redundancy detection', 'xacml policy analysis'])\n", "('NONE', 'an architectural and XXXXX of both XXXXX is presented', ['performance analysis', 'computational models'])\n", "('NONE', 'XXXXX in the atmega 128 of mica2 to compute the XXXXX', ['experimental analysis', 'execution time'])\n", "('method used for task', 'both XXXXX are useful in different contexts for XXXXX', ['hardware accelerators', 'risk analysis'])\n", "('method used for task', 'the XXXXX is found to provide better results than the XXXXX', ['svm classifier', 'pnn classifier'])\n", "('NONE', 'this study is the application of 2d linear XXXXX ( ca ) rules with the help of XXXXX to the problems of edge detection', ['cellular automata', 'fuzzy membership function'])\n", "('NONE', 'this study is the application of 2d linear XXXXX ( ca ) rules with the help of fuzzy membership function to the problems of XXXXX', ['cellular automata', 'edge detection'])\n", "('NONE', 'this study is the application of 2d linear cellular automata ( ca ) rules with the help of XXXXX to the problems of XXXXX', ['fuzzy membership function', 'edge detection'])\n", "('method used for task', 'an efficient and simple thresholding technique of XXXXX based on ca transition rules optimized by XXXXX ( pso ) is proposed', ['edge detection', 'particle swarm optimization method'])\n", "('NONE', 'coded XXXXX out performs uncoded system in terms of XXXXX', ['multi-carrier mimo system', 'ber performance'])\n", "('NONE', 'coded XXXXX with mmse/osic provides better XXXXX than system with coded zf/osic', ['multi-carrier mimo system', 'ber performance'])\n", "('NONE', 'a XXXXX using nsga-ii , entropy and XXXXX is employed', ['two-stage approach', 'topsis methods'])\n", "('NONE', 'a fully automatic system for XXXXX using single-spectral XXXXX is presented', ['tumor segmentation', 'mr images'])\n", "('NONE', 'a study for evaluating the efficacy of XXXXX over XXXXX is included', ['statistical features', 'gabor wavelet features'])\n", "('method used for task', 'randomized ftvbt elects tree nodes based on a XXXXX and increases XXXXX', ['weight function', 'network lifetime'])\n", "('method used for task', 'aecfv exhibits a high XXXXX , low false positive rate , faster attack detection , and lower XXXXX', ['detection rate', 'communication overhead'])\n", "('method used for task', 'data is intelligently analyzed using XXXXX ( hmm ) and XXXXX ( dtw )', ['hidden markov models', 'dynamic time warping'])\n", "('NONE', 'XXXXX determines the severity of the accident and reduces XXXXX', ['data analysis', 'false positives'])\n", "('NONE', 'a major XXXXX in synthesis of XXXXX is state assignment ( sa )', ['optimization problem', 'sequential circuits'])\n", "('NONE', 'a major XXXXX in synthesis of sequential circuits is XXXXX ( sa )', ['optimization problem', 'state assignment'])\n", "('method used for task', 'a major optimization problem in synthesis of XXXXX is XXXXX ( sa )', ['sequential circuits', 'state assignment'])\n", "('method used for task', 'the use of XXXXX microphone array for XXXXX and fusion is explored', ['data capture', 'kinect sensor'])\n", "('NONE', 'the XXXXX specifies the signal XXXXX', ['cost function', 'noise level'])\n", "('NONE', 'gamma equalization for XXXXX can improves the XXXXX', ['contrast enhancement', 'segmentation performance'])\n", "('NONE', 'to facilitate XXXXX , a XXXXX is proposed', ['confusion matrix', 'segmentation performance'])\n", "('method used for task', 'minimizes the XXXXX length , which , in turn , helps to reduce the XXXXX', ['test sequence', 'memory size'])\n", "('NONE', 'usage of a reduced XXXXX minimizes XXXXX during testing period', ['power dissipation', 'test sequence'])\n", "('NONE', 'by analysis of XXXXX of adjacent pixels as well as image interpolation techniques , a novel XXXXX is put forward', ['statistical properties', 'interpolation method'])\n", "('NONE', 'generally speaking , an assessment approach for the XXXXX situation by integrating the XXXXX with data fusion techniques is proposed', ['multi-agent system', 'rock burst'])\n", "('NONE', 'case study results shows the proposed XXXXX situation assessment model can give a relatively accurate forecast , and can help coal mine decision-makers to grasp an overview of the XXXXX of the XXXXX', ['rock burst', 'development trend'])\n", "('NONE', 'case study results shows the proposed XXXXX situation assessment model can give a relatively accurate forecast , and can help coal mine decision-makers to grasp an overview of the XXXXX of the XXXXX', ['development trend', 'rock burst'])\n", "('method used for task', 'the faster implementation of the XXXXX is useful during XXXXX of the 3d bio-medical images', ['affine transform', 'image registration'])\n", "('NONE', 'this work make a trade-off between XXXXX and its reliability associated with XXXXX and energy consumptions', ['modular design', 'fault tolerance'])\n", "('method used for task', 'refat enables run-time reconfiguration of XXXXX and XXXXX', ['network topology', 'routing algorithm'])\n", "('method used for task', 'XXXXX and XXXXX are reduced meanwhile area over-head is kept low', ['execution time', 'power consumption'])\n", "('NONE', 'an XXXXX for lifecycle energy consumption of XXXXX is proposed and evaluated', ['analytical model', 'mobile devices'])\n", "('NONE', 'a comparative XXXXX of performance and XXXXX of old and new android-based smartphone operating under the thin client paradigm was performed', ['empirical evaluation', 'energy consumption'])\n", "('NONE', 'a comparative XXXXX of performance and energy consumption of old and new android-based smartphone operating under the XXXXX was performed', ['empirical evaluation', 'thin client paradigm'])\n", "('NONE', 'a comparative empirical evaluation of performance and XXXXX of old and new android-based smartphone operating under the XXXXX was performed', ['energy consumption', 'thin client paradigm'])\n", "('method used for task', 'an XXXXX is then proposed to solve the XXXXX based on matrix distance', ['efficient algorithm', 'convex problem'])\n", "('NONE', 'framework tested in electrical XXXXX as a XXXXX', ['distribution systems', 'case study'])\n", "('NONE', 'we can do XXXXX with the XXXXX between acf ( t ) and a reference line', ['spectrum sensing', 'euclidean distance'])\n", "('method used for task', 'results show that XXXXX is preserved while XXXXX is enhanced', ['image brightness', 'image contrast'])\n", "('NONE', 'we predict XXXXX of embedded systems to extend their XXXXX', ['power consumption', 'service time'])\n", "('NONE', 'a XXXXX can directly achieve the XXXXX without scanning all channels', ['sensor node', 'l2 handover'])\n", "('method used for task', 'the XXXXX removes the ambiguity and improves the XXXXX', ['correlation function', 'tracking performance'])\n", "('method used for task', 'an XXXXX and a XXXXX of the system are presented', ['energy management', 'control strategy'])\n", "('NONE', 'we propose a game for reducing XXXXX in XXXXX', ['power consumption', 'data centers'])\n", "('method used for task', 'we establish XXXXX on the XXXXX of the game', ['upper bounds', 'convergence time'])\n", "('NONE', 'XXXXX , XXXXX & packet e2e delay are used to evaluate the protocols', ['packet loss', 'packet delivery'])\n", "('method used for task', 'applying weighted XXXXX to classify raw and decomposed XXXXX', ['permutation entropy', 'eeg signals'])\n", "('NONE', 'XXXXX denoising using nlm and XXXXX proposed', ['image sequence', 'zernike moments'])\n", "('method used for task', 'XXXXX is used to attain the more accurate XXXXX', ['sub-pixel accuracy', 'image alignment'])\n", "('NONE', 'XXXXX among different XXXXX', ['performance comparison', 'super-resolution algorithms'])\n", "('NONE', 'the XXXXX of the XXXXX ( cr ) network is improved , when malicious XXXXX user coexists in the network', ['cognitive radio', 'detection performance'])\n", "('NONE', 'the results indicate that for a constant detachment length , the XXXXX and surface effect are linked and it can be seen that the influence of XXXXX decreases with increasing XXXXX', ['beam thickness', 'surface effects'])\n", "('NONE', 'the results indicate that for a constant detachment length , the XXXXX and surface effect are linked and it can be seen that the influence of XXXXX decreases with increasing XXXXX', ['surface effects', 'beam thickness'])\n", "('NONE', 'XXXXX in terms of XXXXX is improved besides memory reduction obtained', ['image quality', 'signal-to-noise ratio'])\n", "('method used for task', 'decreasing XXXXX reduces system complexity and XXXXX', ['processing time', 'power consumption'])\n", "('NONE', 'we increase the XXXXX ratio ( pdr ) through the application of multi-channel technology in XXXXX', ['wireless sensor networks', 'packet delivery'])\n", "('method used for task', 'emd-bpnn has higher XXXXX and better XXXXX than standard bpnn and standard svr', ['prediction accuracy', 'generalization performance'])\n", "('method used for task', 'emd-bpnn can be used as a suitable and effective XXXXX for predicting XXXXX in intensive aquaculture', ['modeling tool', 'water temperature'])\n", "('method used for task', 'two pair-wise XXXXX are proposed for cdma based XXXXX', ['mac protocol', 'code assignment schemes'])\n", "('NONE', 'enhances the security of embedded message by using scene-change detection and discrete XXXXX of XXXXX', ['video sequences', 'cosine transforms'])\n", "('NONE', 'enhances XXXXX using discrete wavelet transforms in XXXXX', ['video quality', 'video sequences'])\n", "('NONE', 'social-spider optimization for XXXXX in XXXXX', ['model selection', 'support vector machines'])\n", "('method used for task', 'minimal added XXXXX needed to expand the score and avoid XXXXX', ['time complexity', 'information loss'])\n", "('method used for task', 'multi-level XXXXX is proposed to enhance XXXXX of lbp', ['discrimination power', 'quantization scheme'])\n", "('method used for task', 'a XXXXX is preprocessed to produce a XXXXX object and a blur image object', ['test image', 'noise image'])\n", "('method used for task', 'a XXXXX is preprocessed to produce a noise XXXXX and a blur XXXXX', ['test image', 'image object'])\n", "('method used for task', 'a test image is preprocessed to produce a XXXXX object and a blur XXXXX', ['noise image', 'image object'])\n", "('method used for task', 'a new energy efficient XXXXX has been proposed for cluster based XXXXX', ['routing algorithm', 'wireless sensor networks'])\n", "('method used for task', 'seed achieves o ( 1 ) message exchange complexity per XXXXX and XXXXX for cluster formation is o ( n ) for wsn having n sensor nodes', ['sensor node', 'time complexity'])\n", "('NONE', 'seed achieves o ( 1 ) XXXXX complexity per XXXXX and time complexity for cluster formation is o ( n ) for wsn having n sensor nodes', ['sensor node', 'message exchange'])\n", "('method used for task', 'seed achieves o ( 1 ) message exchange complexity per XXXXX and time complexity for cluster formation is o ( n ) for wsn having n XXXXX', ['sensor node', 'sensor nodes'])\n", "('method used for task', 'seed achieves o ( 1 ) XXXXX complexity per sensor node and XXXXX for cluster formation is o ( n ) for wsn having n sensor nodes', ['time complexity', 'message exchange'])\n", "('method used for task', 'seed achieves o ( 1 ) message exchange complexity per sensor node and XXXXX for cluster formation is o ( n ) for wsn having n XXXXX', ['time complexity', 'sensor nodes'])\n", "('method used for task', 'seed achieves o ( 1 ) XXXXX complexity per sensor node and time complexity for cluster formation is o ( n ) for wsn having n XXXXX', ['message exchange', 'sensor nodes'])\n", "('NONE', 'the computational accuracy of the propose method in the XXXXX and the overall XXXXX can reach up10−6–10−7', ['gaussian function', 'rbf nn'])\n", "('method used for task', 'the application of the vhdl code for a 3-5-1 XXXXX to the dynamic identification in the XXXXX and the pmsm drive system are successfully demonstrated', ['linear system', 'rbf nn'])\n", "('NONE', 'coded idma system with mutp based on msnr XXXXX regime results in a superior XXXXX with less snr when compared to the equal XXXXX policy', ['power allocation', 'ber performance'])\n", "('method used for task', 'we use a XXXXX to analyze the latency and XXXXX', ['probabilistic model', 'stability performances'])\n", "('NONE', 'the performance of hybrid XXXXX with median , median edge , gradient adjusted and local XXXXX is compared', ['local algorithm', 'prediction algorithms'])\n", "('NONE', 'the hybrid local XXXXX has the highest frequency of zero XXXXX', ['prediction algorithm', 'prediction error'])\n", "('NONE', 'the psnr and XXXXX are improved in the hybrid XXXXX', ['embedding capacity', 'local algorithm'])\n", "('method used for task', 'highlightanalyze the nonlinear cubic XXXXX ratio effect for ultrasonic XXXXX', ['stiffness coefficient', 'cutting system'])\n", "('method used for task', 'ultrasonic XXXXX is designed for stable and synchronized vibration when the excitation force is controlled at 2.0 with the interval of nonlinear cubic XXXXX ratio 0.006 ≦ η < 1.248', ['cutting system', 'stiffness coefficient'])\n", "('method used for task', 'filtering framework is based on XXXXX and XXXXX', ['temporal information', 'sparse representation'])\n", "('NONE', 'after a comprehensive comparison of sparse recovery algorithms , three were selected for our method : bayesian XXXXX ( bcs ) , bregman XXXXX , and orthogonal matching pursuit ( omp )', ['compressive sensing', 'iterative algorithm'])\n", "('method used for task', 'we survey the state scheduling-based XXXXX protocols suitable for unattended XXXXX', ['topology control', 'wireless sensor networks'])\n", "('method used for task', 'a new XXXXX extraction method is proposed based on XXXXX and distance regularized cv ( chan–vese ) model', ['image decomposition', 'river target'])\n", "('NONE', 'the mechanism relies on XXXXX , XXXXX and bit error rate', ['received signal strength', 'congestion level'])\n", "('NONE', 'evaluate the effectiveness of using cellular positioning techniques along with gps-sensor based XXXXX via performing experiments on XXXXX and location accuracy for various data acquisition modes where only gps positioning , only cellular positioning and both of the techniques are executed', ['data collection', 'energy consumption'])\n", "('method used for task', 'evaluate the effectiveness of using cellular positioning techniques along with gps-sensor based XXXXX via performing experiments on energy consumption and XXXXX for various data acquisition modes where only gps positioning , only cellular positioning and both of the techniques are executed', ['data collection', 'location accuracy'])\n", "('method used for task', 'evaluate the effectiveness of using cellular positioning techniques along with gps-sensor based data collection via performing experiments on XXXXX and XXXXX for various data acquisition modes where only gps positioning , only cellular positioning and both of the techniques are executed', ['energy consumption', 'location accuracy'])\n", "('NONE', 'our proposed system is designed to be used for average velocity calculation which is an essential parameter in traffic monitoring systems . therefore , we perform experiments to compute the average velocity by extracting XXXXX with the hybrid data acquisition model . we assume average speed measurements obtained from gps positioning as the XXXXX to assess the performance of our model', ['location data', 'ground truth data'])\n", "('NONE', 'assess the feasibility of cellular positioning in cases where the gps sensors are not available for location estimation . in this paper , we propose a hybrid acquisition model which applies cellular positioning techniques to obtain the raw XXXXX where gps sensor is not available or battery of a particular XXXXX is too low for location lookup', ['smart phone', 'location data'])\n", "('method used for task', 'when a XXXXX round is finished , an adaptive function is carried out to adjust the criteria evaluation in order to obtain a more adaptive ranking order of candidate XXXXX for the next round', ['data exchange', 'mobile devices'])\n", "('method used for task', 'when a XXXXX round is finished , an adaptive function is carried out to adjust the XXXXX in order to obtain a more adaptive ranking order of candidate mobile devices for the next round', ['data exchange', 'criteria evaluation'])\n", "('NONE', 'when a data exchange round is finished , an adaptive function is carried out to adjust the XXXXX in order to obtain a more adaptive ranking order of candidate XXXXX for the next round', ['mobile devices', 'criteria evaluation'])\n", "('method used for task', 'an XXXXX for routing XXXXX using recursive partition is proposed', ['efficient algorithm', 'multicast traffic'])\n", "('method used for task', 'a new XXXXX model to evaluate security risks associated with XXXXX on social pervasive applications', ['information sharing', 'risk indicator'])\n", "('NONE', 'a new risk indicator model to evaluate XXXXX associated with XXXXX on social pervasive applications', ['information sharing', 'security risks'])\n", "('method used for task', 'a new XXXXX model to evaluate XXXXX associated with information sharing on social pervasive applications', ['risk indicator', 'security risks'])\n", "('NONE', 'algorithm aims to improve the joint XXXXX in XXXXX based wban applications', ['signal recovery', 'compressed sensing'])\n", "('NONE', 'the XXXXX of sensing time and XXXXX of sus is achieved through novel iterative dinkelbach method ( nidm ) algorithm', ['joint optimization', 'transmission power'])\n", "('method used for task', 'the proposed XXXXX is compared with standard technologies and recent protocols in the literature , and it achieves better results for XXXXX , packet loss ratio , and throughput parameters', ['mac protocol', 'end-to-end delay'])\n", "('NONE', 'model the XXXXX by means of a XXXXX', ['context evolution', 'graph approach'])\n", "('method used for task', 'a XXXXX is proposed using the XXXXX', ['harmony search algorithm', 'feature selection approach'])\n", "('NONE', 'table-based routing can be used to implement XXXXX by reconfiguring table entries . the article shows how table entries can be computed efficiently , and how the XXXXX can be organized to function reliably even in presence of transmission errors', ['fault-tolerant routing', 'reconfiguration process'])\n", "('NONE', 'the additional hardware overhead for XXXXX table reconfiguration amounts to only 6 % of the XXXXX of a network switch', ['fault-tolerant routing', 'chip area'])\n", "('NONE', 'the coding XXXXX reaches 34 % , compared to the case of coding with no XXXXX', ['performance improvement', 'background suppression'])\n", "('NONE', 'detailed XXXXX of XXXXX , seig , and nonlinear load are presented', ['mathematical model', 'wind turbine'])\n", "('method used for task', 'research highlightsa multi-step XXXXX based on adaptive XXXXX generalized gradient vector flow using component-based normalization for snake model is proposed', ['decision model', 'edge preserving'])\n", "('method used for task', 'research highlightsa multi-step XXXXX based on adaptive edge preserving generalized gradient vector flow using component-based normalization for XXXXX is proposed', ['decision model', 'snake model'])\n", "('method used for task', 'research highlightsa multi-step decision model based on adaptive XXXXX generalized gradient vector flow using component-based normalization for XXXXX is proposed', ['edge preserving', 'snake model'])\n", "('NONE', 'the proposed algorithm presents a novel external force , which provides better results than other approaches in terms of XXXXX , weak XXXXX and convergence', ['noise robustness', 'edge preserving'])\n", "('method used for task', 'an improved multi-step XXXXX based on this novel external force is adopted , which adds new effective weighting function to attenuate the magnitudes of unwanted edges and adopts narrow band method to reduce XXXXX', ['decision model', 'time complexity'])\n", "('method used for task', 'the surface estimation problem is described as a max-margin based formulation of a XXXXX and solving the XXXXX using sub-gradient method', ['kernel function', 'objective function'])\n", "('NONE', 'the XXXXX problem is described as a max-margin based formulation of a XXXXX and solving the objective function using sub-gradient method', ['kernel function', 'surface estimation'])\n", "('method used for task', 'the XXXXX problem is described as a max-margin based formulation of a kernel function and solving the XXXXX using sub-gradient method', ['objective function', 'surface estimation'])\n", "('NONE', 'open issues and XXXXX of metaheuristics for XXXXX are addressed', ['future trends', 'healthcare system'])\n", "('NONE', 'a new modified grey level co-occurrence matrix ( mglcm ) method is presented to extract XXXXX statistical XXXXX for discriminating brain abnormality', ['second order', 'texture features'])\n", "('method used for task', 'mglcm generates efficient XXXXX that is used for measuring the symmetry of XXXXX scan than the tradition glcm', ['texture feature', 'mri brain'])\n", "('NONE', 'the XXXXX of XXXXX are optimized by genetic algorithm', ['membership functions', 'fuzzy controller'])\n", "('NONE', 'the XXXXX of fuzzy controller are optimized by XXXXX', ['membership functions', 'genetic algorithm'])\n", "('NONE', 'the membership functions of XXXXX are optimized by XXXXX', ['fuzzy controller', 'genetic algorithm'])\n", "('NONE', 'the proposed algorithm improves XXXXX even with long XXXXX', ['convergence speed', 'adaptive filters'])\n", "('NONE', 'the led lighting strip is controlled by the XXXXX via a wi-fi XXXXX', ['mobile app', 'wireless network'])\n", "('NONE', 'the XXXXX was performed by using XXXXX ( svm ) and k nearest neighbors ( knn )', ['emotion classification', 'support vector machine'])\n", "('method used for task', 'sensitivity and XXXXX for gis-multicriteria XXXXX is underdeveloped in giscience', ['uncertainty analysis', 'decision making'])\n", "('method used for task', 'XXXXX included for XXXXX from matching keypoints', ['outlier removal', 'ransac algorithm'])\n", "('method used for task', 'we propose a simple and effective XXXXX for XXXXX', ['3d meshes', 'partitioning algorithm'])\n", "('NONE', 'XXXXX are estimated accurately using new XXXXX', ['camera parameters', 'energy function'])\n", "('NONE', 'consider the influence of the bond condition on the XXXXX of XXXXX', ['reinforced concrete beams', 'fire resistance'])\n", "('NONE', 'assess the integrity of XXXXX under XXXXX', ['reinforced concrete beams', 'fire conditions'])\n", "('method used for task', 'XXXXX and XXXXX of local beam sections', ['numerical analysis', 'parametric study'])\n", "('NONE', 'simulation of complex XXXXX with non-homogeneous XXXXX', ['flow problems', 'material parameters'])\n", "('NONE', 'a XXXXX for analysis of virtual XXXXX is developed', ['conceptual framework', 'supply chains'])\n", "('method used for task', 'development of the agribusiness XXXXX based on analytic XXXXX', ['location model', 'hierarchy process'])\n", "('NONE', 'XXXXX with exponential weight-criterion XXXXX and constraints', ['genetic algorithm', 'fitness function'])\n", "('method used for task', 'the XXXXX is XXXXX and easily wearable', ['hybrid system', 'low cost'])\n", "('method used for task', 'results of XXXXX are compared to results of various XXXXX', ['oxygen saturation', 'clinical trials'])\n", "('NONE', 'we design patient-specific XXXXX shapes using XXXXX', ['topology optimization', 'bone replacement'])\n", "('NONE', 'the algorithm denoise XXXXX recorded in noisy environments by XXXXX', ['mobile devices', 'pcg signals'])\n", "('NONE', 'planar 2d XXXXX are in good agreement with complex XXXXX', ['numerical models', '3d models'])\n", "('method used for task', 'presenting a new two-stage meta-heuristic XXXXX based on general XXXXX', ['clustering algorithm', 'type-2 fuzzy sets'])\n", "('NONE', 'incorporating a new similarity-based XXXXX using alpha-plane representation of general XXXXX', ['objective function', 'type-2 fuzzy sets'])\n", "('NONE', 'XXXXX about 8-ms for detection of the XXXXX', ['response time', 'qrs complexes'])\n", "('method used for task', 'classifiers methodologies used included XXXXX and lasso XXXXX', ['genetic algorithms', 'logistic regression'])\n", "('NONE', 'entropies , hos , fd and XXXXX are extracted from XXXXX', ['fundus images', 'gabor wavelet features'])\n", "('method used for task', 'XXXXX provide an opportunity to improve XXXXX', ['electronic medical records', 'patient care'])\n", "('NONE', 'XXXXX : XXXXX and fuzzy connectedness is employed', ['soft computing', 'evolutionary computation'])\n", "('method used for task', 'XXXXX : evolutionary computation and XXXXX is employed', ['soft computing', 'fuzzy connectedness'])\n", "('method used for task', 'soft computing : XXXXX and XXXXX is employed', ['evolutionary computation', 'fuzzy connectedness'])\n", "('NONE', 'model discovery by large scale XXXXX coupled with XXXXX', ['feature extraction', 'machine learning'])\n", "('method used for task', 'a XXXXX for glucose utilization is proposed based on XXXXX', ['global model', 'time-varying parameters'])\n", "('method used for task', 'patient-specific XXXXX for XXXXX are proposed to avoid leakage', ['graph cuts', 'shape constraints'])\n", "('NONE', 'analysis of XXXXX in 62 consecutive patients with XXXXX ( af )', ['atrial fibrillation', 'surface ecg'])\n", "('NONE', 'in this paper , bel is applied for the XXXXX of gene-expression XXXXX', ['microarray data', 'classification tasks'])\n", "('NONE', 'proposed method improves the XXXXX of srbct , hgg and XXXXX', ['detection accuracy', 'lung cancer'])\n", "('method used for task', 'we calculate the XXXXX between the adjusted endpoints to find the accurate XXXXX', ['shortest path', 'splitting line'])\n", "('method used for task', 'we present an analysis of XXXXX and XXXXX applied to odor discrimination', ['pattern recognition', 'dimensionality reduction techniques'])\n", "('method used for task', 'proposing a XXXXX for pediatric XXXXX', ['decision support system', 'epilepsy diagnosis'])\n", "('NONE', 'evaluation of graph theory measures of XXXXX in pediatric XXXXX', ['brain functional connectivity', 'epilepsy diagnosis'])\n", "('method used for task', 'XXXXX to investigate XXXXX and tracking convergence', ['lyapunov stability theorem', 'global stability'])\n", "('method used for task', 'many gene selection methods with XXXXX and XXXXX increase the accuracy of autism recognition', ['genetic algorithm', 'svm classifiers'])\n", "('NONE', 'XXXXX can measure complexity of clinical XXXXX', ['permutation entropy', 'time series'])\n", "('method used for task', 'we compared XXXXX to XXXXX', ['logistic regression models', 'data mining algorithms'])\n", "('NONE', 'the XXXXX performed as well as a simple XXXXX', ['logistic regression model', 'data mining algorithms'])\n", "('NONE', 'application 2 : investigation of XXXXX ( XXXXX of the action potentials along the fibres )', ['muscle activity', 'conduction velocity'])\n", "('NONE', 'application 2 : investigation of XXXXX ( conduction velocity of the XXXXX along the fibres )', ['muscle activity', 'action potentials'])\n", "('NONE', 'application 2 : investigation of muscle activity ( XXXXX of the XXXXX along the fibres )', ['conduction velocity', 'action potentials'])\n", "('NONE', 'a XXXXX of the XXXXX is presented', ['parametric model', 'left ventricle'])\n", "('method used for task', 'this is the first study to validate skin XXXXX to a XXXXX', ['gold standard', 'thickness estimation'])\n", "('NONE', 'XXXXX resolves XXXXX in simulation of neural systems', ['discontinuous galerkin fem', 'stability problem'])\n", "('NONE', 'XXXXX excels XXXXX for lif and lifb models', ['monte carlo simulation', 'discontinuous galerkin fem'])\n", "('NONE', 'a system-level XXXXX of chronic high XXXXX effects is conducted', ['computer simulation', 'sodium intake'])\n", "('method used for task', 'modeling results suggest high XXXXX alters the body-fluid homeostasis and increases the lv XXXXX and relative wall thickness', ['sodium intake', 'work load'])\n", "('method used for task', 'modeling results suggest high XXXXX alters the body-fluid homeostasis and increases the lv work load and relative XXXXX', ['sodium intake', 'wall thickness'])\n", "('method used for task', 'modeling results suggest high sodium intake alters the body-fluid homeostasis and increases the lv XXXXX and relative XXXXX', ['work load', 'wall thickness'])\n", "('method used for task', 'first implementation of a XXXXX with dendritic processing to solve classify XXXXX', ['retinal images', 'lattice neural network'])\n", "('NONE', 'end-tidal XXXXX decreased when a fall in XXXXX diagnosed', ['carbon dioxide', 'cardiac output'])\n", "('NONE', 'multiple expert readers produced a XXXXX of 430 abstracts on XXXXX', ['gold standard', 'lung cancer'])\n", "('NONE', 'we determine the value of XXXXX using 3 disparate XXXXX', ['free text', 'use cases'])\n", "('method used for task', 'XXXXX specific values and limitations are identified in XXXXX', ['use case', 'large data sets'])\n", "('NONE', 'XXXXX are represented by a statistical function with XXXXX', ['normal distribution', 'tissue properties'])\n", "('NONE', 'mean value of the XXXXX are identified using XXXXX', ['genetic algorithm', 'material parameters'])\n", "('NONE', 'using forward XXXXX with XXXXX', ['feature selection', 'svm classification'])\n", "('NONE', 'the state-of-the-art research in XXXXX of XXXXX is presented', ['automated diagnosis', 'celiac disease'])\n", "('NONE', 'we analyze the XXXXX through the XXXXX', ['load distribution', 'wrist joint'])\n", "('method used for task', 'we used the XXXXX model method on a XXXXX of the wrist joint', ['3d model', 'rigid body spring'])\n", "('NONE', 'we used the rigid body spring model method on a XXXXX of the XXXXX', ['3d model', 'wrist joint'])\n", "('NONE', 'we used the XXXXX model method on a 3d model of the XXXXX', ['rigid body spring', 'wrist joint'])\n", "('NONE', 'XXXXX distinguish the three types of XXXXX', ['statistical features', 'eye movements'])\n", "('NONE', 'the proposed XXXXX show superior XXXXX compared to traditional features', ['geometric features', 'classification performance'])\n", "('NONE', 'a novel method for XXXXX of XXXXX ( af ) is proposed', ['automatic detection', 'atrial fibrillation'])\n", "('NONE', 'linking intrinsic features of XXXXX with the optimal XXXXX that maximizes the similarity entropy', ['time series', 'pattern size'])\n", "('NONE', 'the method is built upon XXXXX of XXXXX', ['information fusion', 'endpoint responses'])\n", "('NONE', 'XXXXX trained and tested on a large sample of real XXXXX', ['prediction algorithm', 'clinical data'])\n", "('NONE', 'we applied a XXXXX algorithm-based XXXXX', ['machine learning', 'classification method'])\n", "('method used for task', 'introduced model could be applied in XXXXX and XXXXX', ['ischemia monitoring', 'tumor detection'])\n", "('NONE', 'the proposed system achieves high XXXXX in an XXXXX', ['detection accuracy', 'empirical study'])\n", "('NONE', 'the XXXXX incorporates the edges obtained using a tunable XXXXX', ['cost function', 'gabor filter'])\n", "('NONE', 'the XXXXX depends on XXXXX of matrix fibrous component', ['topological properties', 'healing process'])\n", "('method used for task', 'we propose to accelerate pls–rfe based XXXXX to classify XXXXX', ['gene selection', 'microarray data'])\n", "('method used for task', 'we developed an XXXXX to automatically analyze XXXXX', ['image processing pipeline', 'muscle fibers'])\n", "('NONE', '3d-rsd is fast and more accurate than three other reference methods , and can be used in high-content screening tasks to evaluate XXXXX in human XXXXX lines', ['drug efficacy', 'cancer cell'])\n", "('NONE', 'automatic XXXXX from XXXXX', ['feature extraction', 'retinal images'])\n", "('method used for task', 'XXXXX is common after XXXXX and it is a serious health problem worldwide', ['ventricular tachycardia', 'myocardial infarction'])\n", "('method used for task', 'XXXXX is common after myocardial infarction and it is a serious XXXXX worldwide', ['ventricular tachycardia', 'health problem'])\n", "('method used for task', 'ventricular tachycardia is common after XXXXX and it is a serious XXXXX worldwide', ['myocardial infarction', 'health problem'])\n", "('method used for task', 'the source of reentrant XXXXX is often located in the infarct XXXXX , the layer of thin surviving myocardium adjacent to the infarct', ['ventricular tachycardia', 'border zone'])\n", "('NONE', 'although there is electrical activity occurring in the infarct XXXXX , the XXXXX of the leading edge of the propagating wavefront is not constant', ['conduction velocity', 'border zone'])\n", "('method used for task', 'in this study , the activation rate and ibz XXXXX necessary for functional block to form during premature stimulation and during reentrant XXXXX , are determined', ['geometric structure', 'ventricular tachycardia'])\n", "('method used for task', 'in this study , the XXXXX and ibz XXXXX necessary for functional block to form during premature stimulation and during reentrant ventricular tachycardia , are determined', ['geometric structure', 'activation rate'])\n", "('method used for task', 'in this study , the XXXXX and ibz geometric structure necessary for functional block to form during premature stimulation and during reentrant XXXXX , are determined', ['ventricular tachycardia', 'activation rate'])\n", "('NONE', 'the XXXXX manipulates the cardiac XXXXX to suppress alternans', ['control algorithm', 'tissue mechanics'])\n", "('method used for task', 'improving the reasoning in XXXXX for better XXXXX', ['conceptual graphs', 'knowledge representation'])\n", "('NONE', 'XXXXX of age-related macular degeneration ( amd ) using XXXXX', ['automated detection', 'fundus images'])\n", "('NONE', 'an XXXXX angiogenic process is modeled by a suitable version of the cellular XXXXX', ['in vivo', 'potts model'])\n", "('NONE', 'by detecting associations with XXXXX , XXXXX can be discovered', ['celiac disease', 'research trends'])\n", "('NONE', 'XXXXX were mainly enriched in XXXXX and ribosome pathway', ['feature genes', 'cell proliferation'])\n", "('NONE', 'XXXXX with multimodal data can accurately predict postsurgical outcome in patients with drug resistant mesial XXXXX', ['machine learning', 'temporal lobe epilepsy'])\n", "('method used for task', 'a XXXXX based on the quantitative XXXXX was developed to classify breast masses', ['computer-aided diagnosis system', 'strain features'])\n", "('method used for task', 'we used the combination of XXXXX ( svm ) and improved XXXXX ( iaco )', ['support vector machine', 'ant colony optimization'])\n", "('NONE', 'sbdne is applied to XXXXX using XXXXX', ['cancer classification', 'gene expression data'])\n", "('NONE', 'exploration of 14 XXXXX ( 86 XXXXX )', ['color spaces', 'color features'])\n", "('NONE', 'XXXXX of XXXXX for disease gene and biomarker discovery', ['automated analysis', 'high throughput data'])\n", "('method used for task', 'a novel computer-aided XXXXX for XXXXX using retinal fundus images is proposed', ['decision support system', 'diabetic retinopathy'])\n", "('NONE', 'a novel computer-aided XXXXX for diabetic retinopathy using retinal XXXXX is proposed', ['decision support system', 'fundus images'])\n", "('NONE', 'a novel computer-aided decision support system for XXXXX using retinal XXXXX is proposed', ['diabetic retinopathy', 'fundus images'])\n", "('method used for task', 'ten XXXXX are compared using stability and XXXXX', ['similarity measures', 'feature selection methods'])\n", "('method used for task', 'fs yielding the highest XXXXX are mrmr and XXXXX', ['bhattacharyya distance', 'prediction performance'])\n", "('method used for task', 'the proposed XXXXX is used to control a real three joints robot arm to grasp a XXXXX', ['bci system', 'target object'])\n", "('NONE', 'the controllability of our bci system is improved with the XXXXX embedded in the XXXXX', ['feedback mechanism', 'robot arm'])\n", "('method used for task', 'the controllability of our XXXXX is improved with the XXXXX embedded in the robot arm', ['feedback mechanism', 'bci system'])\n", "('method used for task', 'the controllability of our XXXXX is improved with the feedback mechanism embedded in the XXXXX', ['robot arm', 'bci system'])\n", "('method used for task', 'XXXXX is common after XXXXX', ['ventricular tachycardia', 'myocardial infarction'])\n", "('method used for task', 'reentrant XXXXX is often located in the infarct XXXXX', ['ventricular tachycardia', 'border zone'])\n", "('NONE', 'activation rate and ibz XXXXX during reentrant XXXXX are determined', ['geometric structure', 'ventricular tachycardia'])\n", "('method used for task', 'XXXXX and ibz XXXXX during reentrant ventricular tachycardia are determined', ['geometric structure', 'activation rate'])\n", "('method used for task', 'XXXXX and ibz geometric structure during reentrant XXXXX are determined', ['ventricular tachycardia', 'activation rate'])\n", "('NONE', 'we show how to store intrinsic features from XXXXX in a XXXXX', ['medical images', 'data warehouse'])\n", "('NONE', 'an improved XXXXX based thalamic XXXXX was proposed', ['fuzzy connectedness', 'segmentation method'])\n", "('method used for task', 'investigating a broad range of XXXXX for XXXXX', ['machine learning techniques', 'statistical analyses'])\n", "('NONE', 'evaluation of XXXXX from geometry independent of XXXXX', ['material properties', 'bone strength'])\n", "('method used for task', 'its components are implemented as XXXXX and supported by XXXXX', ['web services', 'cloud services'])\n", "('NONE', 'XXXXX as XXXXX to leverage statistics of the lesion', ['mutual information', 'energy function'])\n", "('NONE', 'three major registration frameworks were examined : ( a ) intensity-based , exhaustive XXXXX using three distinct XXXXX ( b ) geometry-based XXXXX , featuring three geometrical descriptors ( c ) the original implementation of the iterative closest point algorithm', ['cost functions', 'registration framework'])\n", "('NONE', 'three major registration frameworks were examined : ( a ) intensity-based , exhaustive XXXXX using three distinct XXXXX ( b ) geometry-based XXXXX , featuring three geometrical descriptors ( c ) the original implementation of the iterative closest point algorithm', ['cost functions', 'registration framework'])\n", "('method used for task', 'use of geometrical XXXXX for aligning 3d XXXXX', ['feature descriptors', 'medical data'])\n", "('method used for task', 'the devised XXXXX is validated with XXXXX', ['finite element analysis', 'force system'])\n", "('NONE', 'the XXXXX uses a bio-inspired XXXXX on the hermite transform to code local image features', ['optical flow estimation', 'model based'])\n", "('method used for task', 'the XXXXX uses a bio-inspired model based on the hermite transform to code local XXXXX', ['optical flow estimation', 'image features'])\n", "('method used for task', 'the optical flow estimation uses a bio-inspired XXXXX on the hermite transform to code local XXXXX', ['model based', 'image features'])\n", "('NONE', 'the XXXXX allows an unprecedented XXXXX in cell dynamics', ['wavelet transform', 'multi-scale analysis'])\n", "('method used for task', 'we used XXXXX and XXXXX for wheeze recognition', ['short-time fourier transform', 'svm classifier'])\n", "('NONE', 'saccadic XXXXX could be detected efficiently by an XXXXX', ['eye movements', 'adaptive algorithm'])\n", "('method used for task', 'prediction with XXXXX is comparable to multivariate XXXXX', ['random forest', 'linear regression'])\n", "('method used for task', 'the textural XXXXX is extracted by using discrete XXXXX and fractal technology', ['feature set', 'curvelet transform'])\n", "('method used for task', 'the ga–svm method is proposed for XXXXX and wce XXXXX', ['feature selection', 'image classification'])\n", "('NONE', 'XXXXX also reduces the peak von mises stress in the XXXXX and porous cage', ['cortical bone', 'bone fusion'])\n", "('method used for task', 'we proposed a novel gap-search XXXXX ( mrf ) for accurate cervical smear XXXXX', ['markov random field', 'image segmentation'])\n", "('method used for task', 'the method outperforms XXXXX and XXXXX', ['uniform sampling', 'compressive sensing'])\n", "('NONE', 'XXXXX with XXXXX and adaptive genetic operators is proposed', ['hybrid algorithm', 'cuckoo search'])\n", "('NONE', 'XXXXX with cuckoo search and adaptive XXXXX is proposed', ['hybrid algorithm', 'genetic operators'])\n", "('method used for task', 'hybrid algorithm with XXXXX and adaptive XXXXX is proposed', ['cuckoo search', 'genetic operators'])\n", "('method used for task', 'XXXXX showed that omc deposition was highly sensitive to particle charge and size and less sensitive to the inhalation XXXXX', ['sensitivity analysis', 'flow rate'])\n", "('NONE', 'XXXXX of whole body spine is segmented using XXXXX', ['ct images', 'active contour method'])\n", "('method used for task', 'fully automated XXXXX for XXXXX is proposed', ['active contour method', 'initialization method'])\n", "('NONE', 'voxel-classification-based segmentation suffers from XXXXX of XXXXX', ['high dimensionality', 'mr images'])\n", "('method used for task', 'a XXXXX for automatic XXXXX in colonoscopy videos is introduced', ['model based method', 'inflammation detection'])\n", "('method used for task', 'the proposed method is suitable for XXXXX and XXXXX of high-resolution colonoscopy videos', ['parallel implementation', 'real-time processing'])\n", "('NONE', 'XXXXX and state-of-the XXXXX allowed accurate segmentation', ['texture descriptors', 'art features'])\n", "('method used for task', 'a cs framework of XXXXX is proposed for XXXXX ( mecg ) signals in eigenspace', ['data reduction', 'multichannel ecg'])\n", "('NONE', 'pca is used to exploit the XXXXX across the channels resulting into sparse XXXXX', ['spatial correlation', 'eigenspace signals'])\n", "('method used for task', 'using the XXXXX ( cs ) approach , the significant eigenspace signals are gone through further XXXXX', ['compressed sensing', 'dimensionality reduction'])\n", "('NONE', 'using the XXXXX ( cs ) approach , the significant XXXXX are gone through further dimensionality reduction', ['compressed sensing', 'eigenspace signals'])\n", "('NONE', 'using the compressed sensing ( cs ) approach , the significant XXXXX are gone through further XXXXX', ['dimensionality reduction', 'eigenspace signals'])\n", "('NONE', 'a feature selection approach for the large number of XXXXX calculated by XXXXX', ['sparse representation', 'dictionary learning'])\n", "('NONE', 'a XXXXX for the large number of XXXXX calculated by dictionary learning', ['sparse representation', 'feature selection approach'])\n", "('NONE', 'a XXXXX for the large number of sparse representation calculated by XXXXX', ['dictionary learning', 'feature selection approach'])\n", "('NONE', 'XXXXX was carried out by using XXXXX', ['automatic segmentation', 'image processing methods'])\n", "('method used for task', 'statistics of gray levels , XXXXX , and XXXXX were computed', ['texture features', 'shape features'])\n", "('NONE', 'XXXXX of normal and amd classes using XXXXX', ['automated detection', 'fundus images'])\n", "('method used for task', 'XXXXX and XXXXX are used for feature extraction', ['radon transform', 'discrete wavelet transform'])\n", "('method used for task', 'XXXXX and discrete wavelet transform are used for XXXXX', ['radon transform', 'feature extraction'])\n", "('method used for task', 'radon transform and XXXXX are used for XXXXX', ['discrete wavelet transform', 'feature extraction'])\n", "('NONE', 'XXXXX with higher XXXXX affect p–j and p–o fits', ['self-presentation messages', 'argument quality'])\n", "('method used for task', 'interfacing with a XXXXX ( vs. a text screen ) augments efforts to establish XXXXX', ['human body', 'common ground'])\n", "('method used for task', 'XXXXX were more pronounced on the XXXXX', ['practice effects', 'memory measures'])\n", "('method used for task', 'fraping is an accepted social norm amongst some XXXXX , yet is frowned upon by XXXXX', ['young adults', 'older adults'])\n", "('method used for task', 'a XXXXX shows that a sealed bid environment is beneficial to a XXXXX', ['computer simulation', 'search engine'])\n", "('method used for task', 'a XXXXX is studied to reduce the XXXXX of candidate ensembles', ['search space', 'selection strategy'])\n", "('NONE', 'XXXXX increases the interest of students in XXXXX', ['virtual environment', '3d modelling'])\n", "('method used for task', 'overall XXXXX ( osr ) was chosen as the metrics for XXXXX', ['performance evaluation', 'success rate'])\n", "('method used for task', 'analysis and novel architecture for new XXXXX for wireless XXXXX', ['industry standard', 'power transfer'])\n", "('method used for task', 'a fast XXXXX ( mst ) -based clustering methodology that is robust to density-based outliers is proposed for large XXXXX', ['minimum spanning tree', 'high-dimensional data sets'])\n", "('NONE', 'the performance of the proposed algorithms is demonstrated on a number of small low-dimensional and large XXXXX with respect to several state-of-the-art XXXXX', ['clustering algorithms', 'high-dimensional data sets'])\n", "('NONE', 'problem of causal relations in XXXXX among XXXXX is studied', ['frequency domain', 'time series'])\n", "('NONE', 'problem of XXXXX in XXXXX among time series is studied', ['frequency domain', 'causal relations'])\n", "('NONE', 'problem of XXXXX in frequency domain among XXXXX is studied', ['time series', 'causal relations'])\n", "('NONE', 'the XXXXX of the proposed algorithm is compared with dwt and XXXXX', ['resource utilization', 'kalman filter'])\n", "('NONE', 'the ffm-aukf method provides accurate results in the XXXXX of non-linear XXXXX in presence of maneuver', ['state estimation', 'dynamic systems'])\n", "('method used for task', 'XXXXX for at XXXXX', ['channel equalization', 'fading channels'])\n", "('method used for task', 'awi provides good XXXXX and requires a low XXXXX', ['computational cost', 'reconstruction performance'])\n", "('method used for task', 'our algorithm ensures XXXXX and XXXXX of vector map', ['access control', 'copyright protection'])\n", "('NONE', 'our algorithm ensures XXXXX and copyright protection of XXXXX', ['access control', 'vector map'])\n", "('NONE', 'our algorithm ensures access control and XXXXX of XXXXX', ['copyright protection', 'vector map'])\n", "('method used for task', 'an overview is provided of key XXXXX for group and extended XXXXX', ['sequential monte carlo methods', 'object tracking'])\n", "('NONE', 'dna watermarking is for XXXXX and authentication of a XXXXX', ['copyright protection', 'dna sequence'])\n", "('NONE', 'there are n degrees of freedom for designing the XXXXX by using the XXXXX', ['discrete fourier transform', 'mask coefficients'])\n", "('NONE', 'design of the mask coefficients are formulated as an XXXXX with an XXXXX nonconvex function', ['optimization problem', 'l1 norm'])\n", "('NONE', 'design of the XXXXX are formulated as an XXXXX with an l1 norm nonconvex function', ['optimization problem', 'mask coefficients'])\n", "('NONE', 'design of the XXXXX are formulated as an optimization problem with an XXXXX nonconvex function', ['l1 norm', 'mask coefficients'])\n", "('NONE', 'a characterization of the XXXXX corresponding to weighted orthogonal constrained ica algorithms using symmetric XXXXX weighted unitary mapping is obtained', ['stationary points', 'minimum distance'])\n", "('NONE', 'the monotonic XXXXX of the weighted orthogonal constrained fixed point ica algorithms using symmetric XXXXX weighted unitary mapping for convex contrast function is proved , which is further extended to nonconvex contrast function case', ['convergence property', 'minimum distance'])\n", "('method used for task', 'we present a XXXXX for graph analysis that can be used for automatic XXXXX and descriptor evaluation', ['ranking function', 'feature selection'])\n", "('method used for task', 'the goal is to build an XXXXX where descriptors can be analysed for automatic XXXXX', ['evaluation framework', 'feature selection'])\n", "('NONE', '3-dimensional XXXXX in ocean with XXXXX', ['source localization', 'non-gaussian noise'])\n", "('NONE', 'automatic parameter determination in XXXXX based XXXXX', ['sparse representation', 'face hallucination'])\n", "('method used for task', 'XXXXX is analytically tractable under XXXXX', ['regularization parameter', 'map framework'])\n", "('method used for task', 'this new approach has given two new algorithms for XXXXX and XXXXX', ['noise reduction', 'speech enhancement'])\n", "('NONE', 'feedback plant identification is formulated as a XXXXX based XXXXX', ['cuckoo search', 'optimization problem'])\n", "('method used for task', 'XXXXX shows enhanced identification accuracy for XXXXX', ['simulation study', 'cuckoo search'])\n", "('NONE', 'the proposed XXXXX provides excellent XXXXX on noisy images', ['similarity measure', 'recognition performance'])\n", "('method used for task', 'a distributed XXXXX is designed with time delays and XXXXX', ['fusion filter', 'packet losses'])\n", "('NONE', 'a XXXXX for existence of steady-state XXXXX is given', ['sufficient condition', 'fusion filter'])\n", "('method used for task', 'we employ an XXXXX to initialize the XXXXX', ['expectation-maximization algorithm', 'model parameters'])\n", "('NONE', 'according to random finite set ( rfs ) and information inequality , the paper derives the joint detection and estimation ( jde ) XXXXX of an unresolved target-group using single and XXXXX in the presence of missed detection and clutter', ['error bounds', 'multiple sensors'])\n", "('NONE', 'the XXXXX provide important indications of the XXXXX for existing unresolved target-group jde approaches', ['error bounds', 'performance limitations'])\n", "('method used for task', 'inherent mechanisms of the cell-like XXXXX are utilized to realize an efficient and robust XXXXX', ['p system', 'multi-level thresholding method'])\n", "('NONE', 'evaluation of the XXXXX with analysis , XXXXX and experiment', ['numerical simulation', 'delay network'])\n", "('method used for task', 'we present a fast and XXXXX for XXXXX', ['efficient algorithm', 'image inpainting'])\n", "('NONE', 'we present a method for estimating a time-scale local XXXXX on XXXXX', ['hurst exponent', 'time series'])\n", "('method used for task', 'XXXXX is addressed using XXXXX', ['camera model identification', 'hypothesis testing theory'])\n", "('method used for task', 'a novel XXXXX based on the colour and spatial histograms over the XXXXX', ['similarity measure', 'lattice structure'])\n", "('method used for task', 'there is no study describing 2d XXXXX based on XXXXX in the literature', ['metaheuristic algorithms', 'adaptive filter algorithm'])\n", "('method used for task', 'we propose two novel XXXXX for unsupervised non-gaussian XXXXX', ['feature selection', 'inference frameworks'])\n", "('method used for task', 'determination of the optimal XXXXX to minimize the false-alarm without decreasing XXXXX', ['additive noise', 'detection probability'])\n", "('method used for task', 'XXXXX based on XXXXX and frequency warping operator', ['compressive sensing', 'wavelet analysis'])\n", "('method used for task', 'introduce the XXXXX which is more efficient than the XXXXX to fcm based algorithm', ['mahalanobis distance', 'euclidean distance'])\n", "('method used for task', 'combine the XXXXX and the XXXXX with the fcm algorithm based on kl information', ['mahalanobis distance', 'regularization term'])\n", "('NONE', 'combine the XXXXX and the regularization term with the XXXXX based on kl information', ['mahalanobis distance', 'fcm algorithm'])\n", "('NONE', 'combine the mahalanobis distance and the XXXXX with the XXXXX based on kl information', ['regularization term', 'fcm algorithm'])\n", "('method used for task', 'the XXXXX for fda XXXXX is derived with frequency increment error', ['data model', 'mimo radar'])\n", "('NONE', 'the subspace-based XXXXX of the fda XXXXX is analyzed', ['mimo radar', 'localization performance'])\n", "('NONE', 'XXXXX describing the algorithm behavior in XXXXX are obtained ;', ['steady state', 'model expressions'])\n", "('NONE', 'the new XXXXX with the p-norm XXXXX of the error signal is presented', ['cost function', 'logarithmic transformation'])\n", "('method used for task', 'dihs algorithm for improving the XXXXX and XXXXX', ['solution quality', 'search efficiency'])\n", "('method used for task', 'take-one strategy for a XXXXX to globally XXXXX is analyzed', ['fast convergence', 'optimal solution'])\n", "('method used for task', 'a new XXXXX based on new adaptive XXXXX is proposed', ['filtering process', 'diffusion function'])\n", "('method used for task', 'estimate the XXXXX based on signal inpainting and XXXXX', ['sparse representation', 'voltage fluctuation'])\n", "('NONE', 'develop the XXXXX on the recover error of the XXXXX component', ['upper bound', 'voltage fluctuation'])\n", "('method used for task', 'XXXXX is addressed using XXXXX', ['camera model identification', 'hypothesis testing theory'])\n", "('NONE', 'the msd-optimal XXXXX achieves the fastest XXXXX on every iteration', ['step size', 'convergence rate'])\n", "('method used for task', 'a new XXXXX rbp is presented for XXXXX in this paper', ['image descriptor', 'face recognition'])\n", "('method used for task', 'method based on periodic–chaotic states of the XXXXX for XXXXX', ['duffing oscillator', 'time-frequency analysis'])\n", "('NONE', 'the magnetic XXXXX can control the stroke of the pin accurately without any XXXXX', ['feedback control', 'latch mechanism'])\n", "('NONE', 'a high XXXXX can be achieved by the proposed XXXXX', ['contrast ratio', 'pixel circuit'])\n", "('NONE', 'application of organic transparent electrodes based on XXXXX in XXXXX', ['conducting polymers', 'display devices'])\n", "('NONE', 'ramp XXXXX by XXXXX and temperature is analyzed', ['slope variation', 'image load'])\n", "('NONE', 'XXXXX caused by XXXXX is compensated for by redesigning feedback current path in the driver circuit', ['slope variation', 'image load'])\n", "('method used for task', 'XXXXX caused by image load is compensated for by redesigning feedback current path in the XXXXX', ['slope variation', 'driver circuit'])\n", "('method used for task', 'slope variation caused by XXXXX is compensated for by redesigning feedback current path in the XXXXX', ['image load', 'driver circuit'])\n", "('NONE', 'XXXXX was induced by intermolecular XXXXX from eu ( iii ) ion to viologen', ['energy transfer', 'emission control'])\n", "('NONE', 'a novel 7t1c pixel circuit is proposed for XXXXX , high XXXXX , and low power amoled displays', ['high resolution', 'frame rate'])\n", "('method used for task', 'a novel 7t1c pixel circuit is proposed for XXXXX , high frame rate , and XXXXX amoled displays', ['high resolution', 'low power'])\n", "('method used for task', 'a novel 7t1c XXXXX is proposed for XXXXX , high frame rate , and low power amoled displays', ['high resolution', 'pixel circuit'])\n", "('method used for task', 'a novel 7t1c pixel circuit is proposed for high resolution , high XXXXX , and XXXXX amoled displays', ['frame rate', 'low power'])\n", "('method used for task', 'a novel 7t1c XXXXX is proposed for high resolution , high XXXXX , and low power amoled displays', ['frame rate', 'pixel circuit'])\n", "('method used for task', 'a novel 7t1c XXXXX is proposed for high resolution , high frame rate , and XXXXX amoled displays', ['low power', 'pixel circuit'])\n", "('NONE', 'the results show that a small depth ( XXXXX of 2.3–3.9arc min ) allows faster XXXXX times', ['image search', 'disparity range'])\n", "('NONE', 'the results show that a medium depth ( XXXXX of 4.7–7.0arc min ) makes the XXXXX slower', ['image search', 'disparity range'])\n", "('method used for task', 'increasing XXXXX increases channel resistance and XXXXX', ['contact resistance', 'o2 flow'])\n", "('NONE', 'visible decline in μfe as shorter XXXXX of devices with higher XXXXX', ['channel length', 'o2 flow'])\n", "('NONE', 'XXXXX results in reduced XXXXX in all ages', ['visual performance', 'discomfort glare'])\n", "('method used for task', 'a large XXXXX for building mems devices and for XXXXX', ['surface area', 'electrostatic interaction'])\n", "('method used for task', 'study of emissive XXXXX dependent recombination zone and XXXXX', ['energy transfer', 'layer layout'])\n", "('method used for task', 'study of emissive layer layout dependent XXXXX and XXXXX', ['energy transfer', 'recombination zone'])\n", "('NONE', 'study of emissive XXXXX dependent XXXXX and energy transfer', ['layer layout', 'recombination zone'])\n", "('NONE', 'XXXXX indicate that this phosphor also contains fast decay and slow XXXXX', ['decay graph', 'decay process'])\n", "('method used for task', 'XXXXX is easy to implement and XXXXX can reach to 99.3 %', ['motion detection', 'detection accuracy'])\n", "('method used for task', 'the moiré phenomenon is quantified using the XXXXX and XXXXX', ['contrast ratio', 'standard deviation'])\n", "('NONE', 'the effect of XXXXX on visually induced XXXXX ( vims ) and postural control was studied', ['visual motion', 'motion sickness'])\n", "('method used for task', 'visually induced XXXXX ( vims ) seems to be caused by XXXXX', ['motion sickness', 'visual motion'])\n", "('NONE', 'XXXXX indicate that this phosphor also contains fast decay and slow XXXXX', ['decay graph', 'decay process'])\n", "('NONE', 'the XXXXX of paper–pencil test was lower than tablet XXXXX', ['visual fatigue', 'pc test'])\n", "('method used for task', 'defining two terms of computer icons : XXXXX and XXXXX', ['action icon', 'knowledge icon'])\n", "('method used for task', 'analyzing the applied fields of XXXXX and XXXXX', ['action icon', 'knowledge icon'])\n", "('method used for task', 'factors from XXXXX , disparity energy and XXXXX are considered', ['spatial frequency', 'visual attention'])\n", "('method used for task', 'XXXXX is used for XXXXX', ['data embedding', 'sudoku technique'])\n", "('NONE', 'a reversible data hiding scheme that is based on the XXXXX and can achieve the higher XXXXX', ['embedding capacity', 'sudoku technique'])\n", "('method used for task', 'we suggest a new reversible data hiding algorithm based on the probabilistic XXXXX for XXXXX', ['secret sharing scheme', 'color images'])\n", "('NONE', 'the XXXXX and the task-completion time are inversely related . using a smaller display results in a shorter XXXXX', ['completion time', 'display size'])\n", "('NONE', 'effects of the XXXXX on the propagation of XXXXX have been investigated', ['system parameters', 'pressure waves'])\n", "('NONE', 'a facial one-pot XXXXX at XXXXX for ag2se nanoparticles has been developed', ['low temperature', 'synthesis method'])\n", "('method used for task', 'XXXXX ( field equations coupled with optical gain ) is solved by XXXXX using gpgpu card to reduce computation time and to increase accuracy by reducing integration step', ['numerical model', 'parallel computing'])\n", "('method used for task', 'XXXXX ( field equations coupled with optical gain ) is solved by parallel computing using gpgpu card to reduce XXXXX and to increase accuracy by reducing integration step', ['numerical model', 'computation time'])\n", "('method used for task', 'numerical model ( field equations coupled with optical gain ) is solved by XXXXX using gpgpu card to reduce XXXXX and to increase accuracy by reducing integration step', ['parallel computing', 'computation time'])\n", "('NONE', 'effective XXXXX in china need care about review format and XXXXX', ['voting system', 'review systems'])\n", "('NONE', 'we examine the XXXXX of XXXXX', ['moderating effect', 'product category'])\n", "('NONE', 'XXXXX capability plays a mediating role through which business achieves XXXXX', ['competitive advantage', 'business integration'])\n", "('NONE', 'we provided a XXXXX to illustrate the applicability of our XXXXX and our architecture in real life', ['case study', 'trust model'])\n", "('NONE', 'XXXXX and experimental results have shown the efficacy of the proposed XXXXX', ['theoretical analysis', 'incentive mechanism'])\n", "('method used for task', 'relationships between XXXXX and a XXXXX are moderated by negative valence of voc', ['customer expectations', 'brand community interaction'])\n", "('method used for task', 'the positive relationship between XXXXX and XXXXX is moderated by independent self-construal', ['customer satisfaction', 'customer loyalty'])\n", "('method used for task', 'we developed a XXXXX to depict the intention to self-disclose XXXXX in social commerce environment', ['personal information', 'research model'])\n", "('method used for task', 'the XXXXX relies on literature on XXXXX and communication privacy management theory', ['research model', 'information disclosure intention'])\n", "('method used for task', 'the XXXXX relies on literature on information disclosure intention and XXXXX management theory', ['research model', 'communication privacy'])\n", "('method used for task', 'the research model relies on literature on XXXXX and XXXXX management theory', ['information disclosure intention', 'communication privacy'])\n", "('method used for task', 'perceived enjoyment , perceived apathy and XXXXX positively affects intention to disclose XXXXX', ['perceived usefulness', 'personal information'])\n", "('NONE', 'fairness of XXXXX during XXXXX plays an important role in information disclosure intention', ['information exchange', 'social commerce'])\n", "('NONE', 'fairness of XXXXX during social commerce plays an important role in XXXXX', ['information exchange', 'information disclosure intention'])\n", "('NONE', 'fairness of information exchange during XXXXX plays an important role in XXXXX', ['social commerce', 'information disclosure intention'])\n", "('NONE', 'an inferior XXXXX should introduce a q & a service when the XXXXX difference is small', ['search engine', 'search quality'])\n", "('NONE', 'an inferior XXXXX should improve its XXXXX when the XXXXX difference is large', ['search engine', 'search quality'])\n", "('NONE', 'an inferior XXXXX should improve its XXXXX when the XXXXX difference is large', ['search engine', 'search quality'])\n", "('NONE', 'an inferior search engine should improve its XXXXX when the XXXXX difference is large', ['search quality', 'search difference'])\n", "('NONE', 'an inferior search engine should improve its XXXXX when the XXXXX difference is large', ['search quality', 'search difference'])\n", "('NONE', 'XXXXX confirms the effectiveness and value of our XXXXX', ['user study', 'reputation mechanism'])\n", "('method used for task', 'the relationship between XXXXX and XXXXX is moderated by impulse buying tendencies', ['product involvement', 'purchase intention'])\n", "('NONE', 'XXXXX requires XXXXX between banks and telecom operators', ['mobile payment', 'collective action'])\n", "('method used for task', 'social XXXXX and organizational support theory are used to explain XXXXX', ['knowledge contribution', 'exchange theory'])\n", "('method used for task', 'perceived XXXXX and perceived leader support positively affect XXXXX', ['knowledge contribution', 'community support'])\n", "('NONE', 'perceived community support and perceived XXXXX positively affect XXXXX', ['knowledge contribution', 'leader support'])\n", "('method used for task', 'perceived XXXXX and perceived XXXXX positively affect knowledge contribution', ['community support', 'leader support'])\n", "('method used for task', 'we identify the mediation effects of perceived XXXXX and perceived XXXXX', ['community support', 'leader support'])\n", "('NONE', 'XXXXX of reputation in on line transactions is presented in terms of XXXXX', ['theoretical analysis', 'game theory'])\n", "('NONE', 'peripheral cue moderates elaboration on XXXXX but not XXXXX', ['eye movements', 'purchase intention'])\n", "('method used for task', 'the relationship between XXXXX and XXXXX is more significant in high elaboration with negative cue and in low elaboration with positive cue', ['purchase intention', 'eye movement'])\n", "('NONE', 'this study examined drivers of interdependence and XXXXX in XXXXX in virtual communities', ['network convergence', 'social networks'])\n", "('NONE', 'validating the framework by a XXXXX of technology-based XXXXX in mobile payment ecosystem', ['case study', 'market collusion'])\n", "('method used for task', 'use mapreduce to adapt the serial XXXXX for XXXXX', ['map-matching algorithm', 'cloud computing environment'])\n", "('method used for task', 'ogb managers can leverage platform to enhance XXXXX and XXXXX', ['social capital', 'consumer benefits'])\n", "('NONE', 'active participation mediates the effect of XXXXX on ogb XXXXX', ['social capital', 'consumer benefits'])\n", "('method used for task', 'application of XXXXX to examine substitution effects in XXXXX', ['regression analysis', 'fashion e-commerce'])\n", "('NONE', 'XXXXX of multi-criteria cf XXXXX is improved', ['predictive accuracy', 'recommender systems'])\n", "('method used for task', 'XXXXX is affected by the XXXXX : search/purchase goods', ['channel choice', 'product category'])\n", "('NONE', 'incorporated XXXXX of the topic sentence to the XXXXX of sentiment classification', ['feature set', 'feature space'])\n", "('NONE', 'incorporated XXXXX of the topic sentence to the feature space of XXXXX', ['feature set', 'sentiment classification'])\n", "('NONE', 'incorporated feature set of the topic sentence to the XXXXX of XXXXX', ['feature space', 'sentiment classification'])\n", "('method used for task', 'the paper points out the XXXXX such as developing appropriate standardized service description and registry and proposing new ai based methods to achieve automated generation of service description for XXXXX , selection and composition', ['research directions', 'service discovery'])\n", "('NONE', 'the paper points out the XXXXX such as developing appropriate standardized XXXXX and registry and proposing new ai based methods to achieve automated generation of XXXXX for service discovery , selection and composition', ['research directions', 'service description'])\n", "('NONE', 'the paper points out the XXXXX such as developing appropriate standardized XXXXX and registry and proposing new ai based methods to achieve automated generation of XXXXX for service discovery , selection and composition', ['research directions', 'service description'])\n", "('method used for task', 'the paper points out the research directions such as developing appropriate standardized XXXXX and registry and proposing new ai based methods to achieve automated generation of XXXXX for XXXXX , selection and composition', ['service discovery', 'service description'])\n", "('method used for task', 'the paper points out the research directions such as developing appropriate standardized XXXXX and registry and proposing new ai based methods to achieve automated generation of XXXXX for XXXXX , selection and composition', ['service discovery', 'service description'])\n", "('NONE', 'trust can mediate the effects of perceived personalization and XXXXX on XXXXX', ['privacy concerns', 'acceptance intention'])\n", "('NONE', 'for elder potential users , XXXXX have no influences on XXXXX', ['privacy concerns', 'acceptance intention'])\n", "('NONE', 'empirical discovery of the XXXXX of group-level XXXXX', ['mixed effects', 'social capital'])\n", "('method used for task', 'XXXXX and experience affect XXXXX and intentions', ['tie strength', 'message credibility'])\n", "('method used for task', 'we model users’ mobile XXXXX based on a multi-state XXXXX', ['online behavior', 'markov model'])\n", "('method used for task', 'we model users’ mobile XXXXX based on a XXXXX', ['online behavior', 'hidden markov model'])\n", "('NONE', 'typical XXXXX of mobile XXXXX are discovered and analyzed', ['sequential patterns', 'online behavior'])\n", "('method used for task', 'we use bayesian XXXXX to estimate the XXXXX of qoe , considering available prior quality of service ( qos ) data and user ratings', ['data analysis', 'posterior distribution'])\n", "('NONE', 'this is the first study examining XXXXX in XXXXX', ['gender differences', 'online auctions'])\n", "('NONE', 'males exhibit a higher level of XXXXX in XXXXX', ['social interaction', 'online auctions'])\n", "('method used for task', 'we studied the effect of edm and pmedm parameters on XXXXX , wlt and XXXXX of d2 steel', ['fatigue life', 'heat flux'])\n", "('method used for task', 'analyzing and developing XXXXX for the total XXXXX generated by using fem', ['numerical models', 'heat flux'])\n", "('NONE', 'a method for XXXXX of XXXXX is presented', ['automatic generation', 'water distribution systems'])\n", "('NONE', 'XXXXX of lake and XXXXX is generally inadequate', ['performance assessment', 'marine models'])\n", "('NONE', 'designed to allow flexibility in the approach to construct different XXXXX without compiling XXXXX', ['crop models', 'source code'])\n", "('NONE', 'location tracking through time of XXXXX output for ease of XXXXX', ['simulation model', 'decision making'])\n", "('method used for task', 'we review XXXXX related technologies to manage , transfer and process XXXXX', ['web service', 'big data'])\n", "('method used for task', 'a novel muscl scheme is developed to provide more efficient and accurate XXXXX to the 2d XXXXX', ['numerical solutions', 'shallow water equations'])\n", "('NONE', 'XXXXX showed that increasing tree height and local XXXXX were the main factors associated with damage', ['statistical analysis', 'wind speed'])\n", "('NONE', 'XXXXX automatically scripts manual XXXXX data edits in python', ['quality control', 'odm tools'])\n", "('NONE', 'XXXXX preserves provenance of XXXXX edits', ['quality control', 'odm tools'])\n", "('method used for task', 'XXXXX is XXXXX and cross platform compatible', ['open source', 'odm tools'])\n", "('method used for task', 'uncertainties on XXXXX and XXXXX prevail after 2080', ['sea-level rise', 'climate change scenarios'])\n", "('NONE', 'effectiveness of XXXXX depends on global XXXXX', ['boundary conditions', 'policy strategies'])\n", "('NONE', 'we present an overview of sa and its link to XXXXX , XXXXX and evaluation , robust decision-making', ['uncertainty analysis', 'model calibration'])\n", "('NONE', 'XXXXX was assessed through a XXXXX', ['network structure', 'participatory approach'])\n", "('NONE', 'XXXXX improved XXXXX and model response', ['data integration', 'data quality'])\n", "('NONE', 'XXXXX improved data quality and XXXXX', ['data integration', 'model response'])\n", "('method used for task', 'data integration improved XXXXX and XXXXX', ['data quality', 'model response'])\n", "('method used for task', 'application of XXXXX to co-construct XXXXX of flood risk', ['bayesian networks', 'conceptual model'])\n", "('NONE', 'we developed a model for the XXXXX of XXXXX and maintenance', ['joint optimization', 'process control'])\n", "('NONE', 'potential XXXXX can be obtained from joint spc and XXXXX', ['cost savings', 'maintenance policies'])\n", "('method used for task', 'we discuss their XXXXX and present suitable XXXXX', ['computational complexity', 'solution procedures'])\n", "('method used for task', 'we show the basic XXXXX to be solvable in XXXXX', ['allocation problem', 'polynomial time'])\n", "('NONE', 'we obtain a surrogate duality theorems for semi-definite XXXXX in the face of XXXXX', ['optimization problems', 'data uncertainty'])\n", "('method used for task', 'revise comparative properties of the XXXXX for the XXXXX', ['continuous model', 'discrete model'])\n", "('method used for task', 'the objective of this paper is the research on optimization methodologies for a complete integration of XXXXX , crew scheduling and driver rostering problems so as to develop XXXXX', ['vehicle scheduling', 'efficient algorithms'])\n", "('method used for task', 'we have developed a XXXXX based on XXXXX to solve the problem', ['heuristic approach', 'benders decomposition'])\n", "('NONE', 'results are reported on new XXXXX , derived from well-known XXXXX', ['routing problems', 'benchmark instances'])\n", "('method used for task', 'we develop an XXXXX to solve clusterwise XXXXX', ['incremental algorithm', 'linear regression problems'])\n", "('method used for task', 'we establish a formal link between XXXXX and spectral XXXXX', ['expected utility', 'risk measures'])\n", "('NONE', 'we consider three XXXXX applied in XXXXX', ['portfolio theory', 'quadratic optimization problems'])\n", "('NONE', 'the dm makes XXXXX of identified pareto optimal points and gives XXXXX', ['pairwise comparisons', 'reference points'])\n", "('method used for task', 'we propose XXXXX for the XXXXX of a speciality oils’ company', ['linear models', 'supply chain'])\n", "('method used for task', 'this article provides an XXXXX for a serial XXXXX with multiple suppliers and time varying demand', ['inventory model', 'supply chain'])\n", "('NONE', 'our results show that XXXXX and holding costs can affect XXXXX and order lot sizing decisions', ['supplier selection', 'inventory setup'])\n", "('NONE', 'involve XXXXX , XXXXX and bootstrapping', ['discrete-event simulation', 'robust optimization'])\n", "('NONE', 'to explore XXXXX under XXXXX with consumer returns', ['inventory control problem', 'consignment contract'])\n", "('NONE', 'to explore XXXXX under consignment contract with XXXXX', ['inventory control problem', 'consumer returns'])\n", "('NONE', 'to explore inventory control problem under XXXXX with XXXXX', ['consignment contract', 'consumer returns'])\n", "('method used for task', 'the main findings of the XXXXX is that XXXXX increases the economic order quantity and could serve as a buyer–supplier coordination mechanism', ['literature review', 'trade credit'])\n", "('method used for task', 'the main findings of the XXXXX is that trade credit increases the economic order quantity and could serve as a buyer–supplier XXXXX', ['literature review', 'coordination mechanism'])\n", "('method used for task', 'the main findings of the XXXXX is that trade credit increases the economic XXXXX and could serve as a buyer–supplier coordination mechanism', ['literature review', 'order quantity'])\n", "('NONE', 'the main findings of the literature review is that XXXXX increases the economic order quantity and could serve as a buyer–supplier XXXXX', ['trade credit', 'coordination mechanism'])\n", "('NONE', 'the main findings of the literature review is that XXXXX increases the economic XXXXX and could serve as a buyer–supplier coordination mechanism', ['trade credit', 'order quantity'])\n", "('NONE', 'the main findings of the literature review is that trade credit increases the economic XXXXX and could serve as a buyer–supplier XXXXX', ['coordination mechanism', 'order quantity'])\n", "('NONE', 'we derive a detailed agenda for XXXXX around two core themes : opportunities arising from inside XXXXX and opportunities arising from outside XXXXX', ['future research', 'operations management'])\n", "('NONE', 'we derive a detailed agenda for XXXXX around two core themes : opportunities arising from inside XXXXX and opportunities arising from outside XXXXX', ['future research', 'operations management'])\n", "('NONE', 'cooperation and XXXXX improve ecologic as well as XXXXX', ['vertical integration', 'economic efficiency'])\n", "('NONE', 'we model the XXXXX that deals with products having high XXXXX', ['supply chain', 'demand uncertainty'])\n", "('method used for task', 'we develop the bidirectional XXXXX to coordinate the XXXXX', ['supply chain', 'option contracts'])\n", "('NONE', 'additional XXXXX as well as XXXXX are proposed to significantly speed up overall solution time', ['lower bounds', 'valid inequalities'])\n", "('NONE', 'additional XXXXX as well as valid inequalities are proposed to significantly speed up overall XXXXX', ['lower bounds', 'solution time'])\n", "('method used for task', 'additional lower bounds as well as XXXXX are proposed to significantly speed up overall XXXXX', ['valid inequalities', 'solution time'])\n", "('method used for task', 'proposed an XXXXX for the two- and three-echelon reverse XXXXX', ['analytical model', 'supply chain'])\n", "('method used for task', 'defined the pc recycling XXXXX as foundation for generating the XXXXX', ['supply chain', 'analytical model'])\n", "('NONE', 'the XXXXX of our model is not only robust but also has a desired XXXXX', ['optimal portfolio', 'factor exposure'])\n", "('NONE', 'interrelationship between XXXXX , XXXXX , and ghgs is considered', ['lead time', 'transportation cost'])\n", "('method used for task', 'a novel solution heuristic for the general lotsizing and XXXXX for parallel XXXXX is presented', ['scheduling problem', 'production lines'])\n", "('NONE', 'XXXXX are calculated by an algorithmic approach using XXXXX', ['nash equilibria', 'linear programming'])\n", "('NONE', 'XXXXX increases logarithmically with number of discrete points , XXXXX decreases linearly', ['problem size', 'optimality gap'])\n", "('NONE', 'the above ideas are demonstrated with a XXXXX real world XXXXX', ['large scale', 'case study'])\n", "('method used for task', 'a XXXXX is suggested to represent the new XXXXX', ['mathematical programming model', 'maintenance scheduling problem'])\n", "('NONE', 'to show the applicability of the model and XXXXX , a XXXXX is reported on real data', ['case study', 'solution algorithm'])\n", "('method used for task', 'XXXXX for block-angular problems based on XXXXX', ['interior-point method', 'hybrid approach'])\n", "('NONE', 'XXXXX successfully imputes XXXXX in time series', ['variable neighborhood search', 'missing values'])\n", "('NONE', 'XXXXX successfully imputes missing values in XXXXX', ['variable neighborhood search', 'time series'])\n", "('NONE', 'variable neighborhood search successfully imputes XXXXX in XXXXX', ['missing values', 'time series'])\n", "('method used for task', 'a heuristic XXXXX based on the XXXXX and subgradient methods is provided', ['lagrangian relaxation', 'solution approach'])\n", "('method used for task', 'we solve the XXXXX in short cpu times to XXXXX below 5 %', ['problem instances', 'optimality gaps'])\n", "('method used for task', 'we develop a numerical nonlinear principal XXXXX ( not tractable to solve in XXXXX )', ['closed form', 'agent problem'])\n", "('method used for task', 'XXXXX for XXXXX of markowitz portfolio weights', ['closed-form formulas', 'estimation errors'])\n", "('NONE', 'XXXXX for estimation errors of XXXXX weights', ['closed-form formulas', 'markowitz portfolio'])\n", "('NONE', 'closed-form formulas for XXXXX of XXXXX weights', ['estimation errors', 'markowitz portfolio'])\n", "('NONE', 'efficient bootstrap-based method for XXXXX of XXXXX', ['empirical assessment', 'estimation errors'])\n", "('NONE', 'we identified five XXXXX that close XXXXX . then we gave a review of each model', ['research streams', 'surrogate duality gaps'])\n", "('method used for task', 'multiperiod XXXXX ( msl ) policies for XXXXX', ['service level', 'supply chains'])\n", "('NONE', 'providing XXXXX of decision variables of specific XXXXX', ['explicit solution', 'dynamic models'])\n", "('NONE', 'providing XXXXX of XXXXX of specific dynamic models', ['explicit solution', 'decision variables'])\n", "('NONE', 'providing explicit solution of XXXXX of specific XXXXX', ['dynamic models', 'decision variables'])\n", "('method used for task', 'finding relationship between the dynamics of XXXXX and XXXXX', ['improvement activities', 'model structures'])\n", "('NONE', 'we introduce the first modeling as full truckload pickup and XXXXX with XXXXX', ['time windows', 'delivery problem'])\n", "('NONE', 'a XXXXX with permutation chromosome and XXXXX is proposed', ['genetic algorithm', 'local search'])\n", "('NONE', 'information granulation of XXXXX used in XXXXX', ['linguistic information', 'group decision making'])\n", "('method used for task', 'XXXXX is used to made operational the XXXXX', ['granular computing', 'linguistic information'])\n", "('NONE', 'XXXXX expressed in terms of XXXXX defined as sets', ['linguistic information', 'information granules'])\n", "('method used for task', 'the granulation of the XXXXX is formulated as an XXXXX', ['linguistic terms', 'optimization problem'])\n", "('method used for task', 'XXXXX need to be readjusted to account for XXXXX', ['risk measures', 'default risk'])\n", "('NONE', 'XXXXX compare against the best known XXXXX', ['computational experiments', 'exact solution methods'])\n", "('NONE', 'XXXXX using XXXXX for packing circles is proposed', ['heuristic algorithm', 'linear models'])\n", "('method used for task', 'an inverse mean-deviation XXXXX finds a deviation measure for a given XXXXX', ['optimal portfolio', 'portfolio problem'])\n", "('method used for task', 'an inverse mean-deviation portfolio problem finds a XXXXX for a given XXXXX', ['optimal portfolio', 'deviation measure'])\n", "('NONE', 'an inverse mean-deviation XXXXX finds a XXXXX for a given optimal portfolio', ['portfolio problem', 'deviation measure'])\n", "('NONE', 'necessary and XXXXX for the existence of such a XXXXX are obtained', ['sufficient conditions', 'deviation measure'])\n", "('NONE', 'we consider two-stage XXXXX with arc XXXXX', ['vehicle routing problem', 'time windows'])\n", "('method used for task', 'we propose a mip formulation and a XXXXX based on XXXXX', ['heuristic approach', 'memetic algorithm'])\n", "('method used for task', 'we propose a XXXXX and a XXXXX based on memetic algorithm', ['heuristic approach', 'mip formulation'])\n", "('method used for task', 'we propose a XXXXX and a heuristic approach based on XXXXX', ['memetic algorithm', 'mip formulation'])\n", "('NONE', 'the XXXXX obtains the XXXXX of small and medium-size problems', ['optimal solution', 'mip formulation'])\n", "('NONE', 'the XXXXX obtains good quality solutions in a short XXXXX', ['memetic algorithm', 'computation time'])\n", "('NONE', 'the XXXXX obtains good XXXXX in a short computation time', ['memetic algorithm', 'quality solutions'])\n", "('NONE', 'the memetic algorithm obtains good XXXXX in a short XXXXX', ['computation time', 'quality solutions'])\n", "('NONE', 'XXXXX , under conditions of XXXXX , is reviewed', ['workforce planning', 'market uncertainty'])\n", "('NONE', 'decisions regarding XXXXX , targeting different XXXXX , are discussed', ['service levels', 'workforce shifts'])\n", "('method used for task', 'develop XXXXX to construct tdsvm XXXXX', ['memetic algorithm', 'classification models'])\n", "('method used for task', 'we use XXXXX to characterize the XXXXX', ['stochastic optimal control', 'optimal policies'])\n", "('NONE', 'XXXXX , XXXXX , input requirement functions are considered', ['production function', 'distance functions'])\n", "('method used for task', 'provides a XXXXX and a decomposition-based meta-heuristic XXXXX', ['model formulation', 'solution method'])\n", "('method used for task', 'proposed XXXXX and a hybrid heuristic solution technique for solving a XXXXX variant', ['mathematical model', 'vehicle routing problem'])\n", "('method used for task', 'proposed XXXXX and a hybrid heuristic XXXXX for solving a vehicle routing problem variant', ['mathematical model', 'solution technique'])\n", "('method used for task', 'proposed mathematical model and a hybrid heuristic XXXXX for solving a XXXXX variant', ['vehicle routing problem', 'solution technique'])\n", "('NONE', 'we study a XXXXX with soft XXXXX and stochastic travel times', ['vehicle routing problem', 'time windows'])\n", "('method used for task', 'we propose an XXXXX based on XXXXX and branch-and-price', ['column generation', 'exact solution approach'])\n", "('NONE', 'a new formulation for the time-dependent multi-zone multi-trip XXXXX with XXXXX', ['vehicle routing problem', 'time window'])\n", "('NONE', 'we introduce a law of one price rule for XXXXX of XXXXX', ['shadow prices', 'undesirable outputs'])\n", "('method used for task', 'a near-optimal joint XXXXX is found that considers XXXXX and batching', ['stochastic demand', 'inventory policy'])\n", "('NONE', 'we study the XXXXX with sequence-independent XXXXX', ['setup times', 'two-machine flowshop problem'])\n", "('method used for task', 'we develop an XXXXX for serial XXXXX', ['supply chains', 'optimization procedure'])\n", "('method used for task', 'discrete malliavin XXXXX is introduced to compute greeks ( sensitivity ) of the XXXXX', ['calculus approach', 'option prices'])\n", "('NONE', 'achieve better XXXXX than a commercial stock cutting XXXXX', ['software package', 'solution quality'])\n", "('method used for task', 'control strategies include rts ( regular XXXXX ) assignment and XXXXX', ['time slots', 'rts reservation'])\n", "('method used for task', 'XXXXX improves the unused time slots and XXXXX of rts assignment', ['waiting times', 'rts reservation'])\n", "('method used for task', 'rts reservation improves the unused XXXXX and XXXXX of rts assignment', ['waiting times', 'time slots'])\n", "('NONE', 'XXXXX improves the unused XXXXX and waiting times of rts assignment', ['rts reservation', 'time slots'])\n", "('NONE', 'XXXXX decreases the XXXXX under the cvar', ['capacity uncertainty', 'order quantity'])\n", "('NONE', 'XXXXX may increase or decrease the XXXXX under the var constraint', ['capacity uncertainty', 'order quantity'])\n", "('NONE', 'XXXXX may increase or decrease the order quantity under the XXXXX', ['capacity uncertainty', 'var constraint'])\n", "('NONE', 'capacity uncertainty may increase or decrease the XXXXX under the XXXXX', ['order quantity', 'var constraint'])\n", "('method used for task', 'under the XXXXX , the effect depends on the XXXXX', ['confidence level', 'var constraint'])\n", "('method used for task', 'conduct the XXXXX for the XXXXX', ['time complexity analysis', 'reduction procedures'])\n", "('method used for task', 'XXXXX and XXXXX illustrate the usefulness of theoretical results', ['numerical examples', 'case study'])\n", "('NONE', 'we measure XXXXX through an output-oriented weighted XXXXX', ['additive model', 'revenue inefficiency'])\n", "('NONE', 'the XXXXX can be found by a XXXXX', ['dynamic programming algorithm', 'optimal routing strategy'])\n", "('NONE', 'the XXXXX has a specific XXXXX when the time horizon is finite or infinite', ['optimal strategy', 'threshold structure'])\n", "('NONE', 'the XXXXX has a specific threshold structure when the XXXXX is finite or infinite', ['optimal strategy', 'time horizon'])\n", "('NONE', 'the optimal strategy has a specific XXXXX when the XXXXX is finite or infinite', ['threshold structure', 'time horizon'])\n", "('method used for task', 'the joint XXXXX is characterized in a continuous XXXXX', ['optimal policy', 'inventory review'])\n", "('NONE', 'considers fractionated intensity-modulated XXXXX of XXXXX under breathing motion uncertainty', ['radiation therapy', 'lung cancer'])\n", "('method used for task', 'proposes a method that generalizes XXXXX and adaptive XXXXX', ['robust optimization', 'radiation therapy'])\n", "('NONE', 'isotonic nonparametric XXXXX ( inls ) fits a XXXXX to data', ['least squares', 'step function'])\n", "('NONE', 'free disposal hull ( fdh ) envelops all XXXXX by XXXXX', ['data points', 'step function'])\n", "('method used for task', 'combined XXXXX , meta-heuristic and XXXXX', ['local search', 'mathematical programming'])\n", "('NONE', 'XXXXX of parameters in the XXXXX', ['stress testing', 'markowitz model'])\n", "('method used for task', 'simultaneous XXXXX and XXXXX is formulated in a job shop', ['lot sizing', 'scheduling problem'])\n", "('method used for task', 'simultaneous XXXXX and scheduling problem is formulated in a XXXXX', ['lot sizing', 'job shop'])\n", "('method used for task', 'simultaneous lot sizing and XXXXX is formulated in a XXXXX', ['scheduling problem', 'job shop'])\n", "('method used for task', 'XXXXX , buyer–supplier relationships and XXXXX are considered', ['knowledge sharing', 'production capacities'])\n", "('method used for task', 'negative correlation , although more erratic , improves XXXXX and XXXXX relative to iid', ['lead time', 'inventory performance'])\n", "('NONE', 'conducts a XXXXX to assess the accuracy of the XXXXX', ['simulation study', 'scenario model'])\n", "('method used for task', 'a mixed XXXXX is introduced for a real short sea XXXXX', ['inventory routing problem', 'integer model'])\n", "('method used for task', 'the intm–intn XXXXX is based on XXXXX', ['decision making', 'choice value soft sets'])\n", "('method used for task', 'computing XXXXX is a heavy load when processing XXXXX', ['big data', 'choice value soft sets'])\n", "('NONE', 'we develop a XXXXX without computation of XXXXX', ['pruning method', 'choice value soft sets'])\n", "('NONE', 'we conclude our XXXXX with XXXXX directions', ['future research', 'review paper'])\n", "('method used for task', 'the XXXXX is decomposed into stage efficiencies in XXXXX', ['system efficiency', 'product form'])\n", "('method used for task', 'develops the XXXXX approach to XXXXX', ['inventory balance', 'order forecasting'])\n", "('NONE', 'XXXXX approach outperforms traditional XXXXX approaches', ['inventory balance', 'order forecasting'])\n", "('NONE', 'we discuss in detail an application in the XXXXX of XXXXX', ['strategic planning', 'freight transportation'])\n", "('method used for task', 'XXXXX and XXXXX for electricity generation is analyzed', ['technology selection', 'capacity investment'])\n", "('NONE', 'synthesis of unique context of XXXXX within a coherent XXXXX framework', ['product families', 'equilibrium decision'])\n", "('NONE', 'we present XXXXX with efficient XXXXX in each iteration', ['tabu search algorithm', 'improvement method'])\n", "('method used for task', 'lagragian XXXXX is proposed to obtain the XXXXX', ['relaxation algorithm', 'lower bounds'])\n", "('method used for task', 'it shows that results of XXXXX are very close to the XXXXX', ['tabu search', 'lower bounds'])\n", "('method used for task', 'a generalized multiplicative XXXXX ( gmddf ) is developed to measure XXXXX ( te )', ['directional distance function', 'technical efficiency'])\n", "('NONE', 'quantifies the effectiveness of a firm’s XXXXX under XXXXX', ['production system', 'demand uncertainty'])\n", "('NONE', 'we assess trade-offs between socio-economic and XXXXX of XXXXX', ['environmental impact', 'manure management'])\n", "('method used for task', 'we apply XXXXX to select best compromise XXXXX solutions', ['compromise programming', 'manure management'])\n", "('NONE', 'our procedures generate strong XXXXX and improve the convergence of XXXXX', ['lower bounds', 'branch-and-bound algorithms'])\n", "('method used for task', 'XXXXX and other XXXXX ( psms ) can help address science and technology risk conflicts', ['problem structuring methods', 'issues mapping'])\n", "('method used for task', 'we present a XXXXX to solve the combined XXXXX', ['branch-and-price algorithm', 'optimization problem'])\n", "('NONE', 'show impact of XXXXX on XXXXX', ['capacity constraints', 'optimal policies'])\n", "('NONE', 'ignoring the XXXXX in XXXXX may lead to inadequate decisions', ['supplier selection', 'currency uncertainty'])\n", "('NONE', 'XXXXX on XXXXX showing that significant improvements can be obtained', ['computational study', 'real-world data'])\n", "('method used for task', 'we derive XXXXX for the ( s−1 , s ) XXXXX with backordering', ['structural properties', 'inventory model'])\n", "('NONE', 'XXXXX without making a XXXXX', ['column generation', 'linear programming relaxation'])\n", "('method used for task', 'customised XXXXX to determine when the restricted master problem contains an XXXXX', ['optimality conditions', 'optimal solution'])\n", "('method used for task', 'customised XXXXX to determine when the restricted XXXXX contains an optimal solution', ['optimality conditions', 'master problem'])\n", "('NONE', 'customised optimality conditions to determine when the restricted XXXXX contains an XXXXX', ['optimal solution', 'master problem'])\n", "('NONE', 'develop equilibrium variety and XXXXX strategies in a retailer-led XXXXX', ['supply chain', 'channel structure'])\n", "('method used for task', 'comparisons based on XXXXX and XXXXX', ['objective function value', 'time consumption'])\n", "('method used for task', 'we build XXXXX for XXXXX in foreclosed housing acquisition', ['stochastic models', 'budget allocation'])\n", "('method used for task', 'interactive XXXXX for handling many objective XXXXX', ['evolutionary algorithm', 'optimization problems'])\n", "('method used for task', 'a XXXXX for XXXXX in copper mines is presented', ['stochastic model', 'strategic planning'])\n", "('NONE', 'XXXXX can enable XXXXX in engineering organizations', ['action research', 'technology management'])\n", "('method used for task', 'a new velocity updating strategy is incorporated into XXXXX to improve XXXXX', ['pso algorithm', 'solution quality'])\n", "('NONE', 'depending on the XXXXX , total XXXXX can become negative', ['model parameters', 'agency costs'])\n", "('method used for task', 'we propose a XXXXX for XXXXX', ['evolutionary multiobjective optimization', 'resource allocation model'])\n", "('method used for task', 'we propose a double-sphere XXXXX for XXXXX', ['crowding distance', 'evolutionary multiobjective optimization'])\n", "('method used for task', 'experiments are performed on twelve XXXXX and one XXXXX', ['benchmark problems', 'real-world problem'])\n", "('method used for task', 'XXXXX and evolutionary based XXXXX are implemented', ['local search', 'solution algorithms'])\n", "('method used for task', 'designed a customized XXXXX to solve XXXXX of practical size', ['memetic algorithm', 'problem instances'])\n", "('method used for task', 'propose a XXXXX and XXXXX for the problem', ['mixed integer programming', 'branch-and-bound algorithm'])\n", "('method used for task', 'we build a corresponding XXXXX for XXXXX and project selection at the level of software skills', ['mathematical model', 'resource allocation'])\n", "('NONE', 'continuous-review XXXXX with XXXXX', ['time-varying parameters', 'inventory problem'])\n", "('NONE', 'classification of the XXXXX within the XXXXX', ['goal programming model', 'portfolio selection'])\n", "('method used for task', 'the perspectives of the XXXXX for the XXXXX', ['goal programming', 'portfolio selection'])\n", "('NONE', 'XXXXX for the competitiveness of the XXXXX', ['problem structuring', 'automotive industry'])\n", "('NONE', 'analysis of competitiveness of the XXXXX using bayesian XXXXX', ['automotive industry', 'causal networks'])\n", "('NONE', 'XXXXX using bayesian XXXXX', ['sensitivity analysis', 'causal networks'])\n", "('method used for task', 'uhgs matches or outperforms all current best problem-tailored algorithms on 29 notable XXXXX and 42 XXXXX', ['vrp variants', 'benchmark sets'])\n", "('NONE', 'can handle the large-scale XXXXX with fairly good XXXXX', ['problem instances', 'solution quality'])\n", "('method used for task', 'namely standard XXXXX and the XXXXX', ['quadratic programming', 'quadratic assignment problem'])\n", "('NONE', 'two-level XXXXX under continuous review with XXXXX', ['supply chain', 'uncertain demand'])\n", "('method used for task', 'a necessary and XXXXX is found for generating an XXXXX', ['sufficient condition', 'optimal solution'])\n", "('method used for task', 'sgp problems are difficult , XXXXX to solve for XXXXX', ['np-hard problems', 'global optimality'])\n", "('method used for task', 'an XXXXX for the cm to derive a reliable XXXXX from a pair-wise comparison matrix', ['optimization model', 'priority vector'])\n", "('method used for task', 'an XXXXX for the cm to derive a reliable priority vector from a pair-wise XXXXX', ['optimization model', 'comparison matrix'])\n", "('NONE', 'an optimization model for the cm to derive a reliable XXXXX from a pair-wise XXXXX', ['priority vector', 'comparison matrix'])\n", "('NONE', 'compare cm with other prioritization methods based on two XXXXX : XXXXX and minimum violation', ['performance evaluation', 'euclidean distance'])\n", "('method used for task', 'induce a XXXXX for a pair-wise XXXXX based on the cm', ['consistency index', 'comparison matrix'])\n", "('method used for task', 'the XXXXX are based on optimality arguments and XXXXX', ['linear programming', 'reduction procedures'])\n", "('NONE', 'XXXXX with three sequential mixed and integer XXXXX', ['optimization methods', 'linear programs'])\n", "('method used for task', 'we propose uacor , a unified XXXXX for XXXXX', ['continuous optimization', 'aco algorithm'])\n", "('NONE', 'we characterize the XXXXX with XXXXX of purchasing time', ['rational expectations equilibrium', 'customer choice'])\n", "('NONE', 'we show how information and XXXXX affect the XXXXX', ['demand dynamics', 'rational expectations equilibrium'])\n", "('NONE', 'found XXXXX oppose the logic of XXXXX', ['fuzzy numbers', 'fuzzy set theory'])\n", "('method used for task', 'found XXXXX can not give an acceptable method to rank XXXXX', ['fuzzy ahp', 'fuzzy numbers'])\n", "('method used for task', 'the heuristics are tested using real-world XXXXX and XXXXX where a congestion charge scheme is in operation', ['road networks', 'traffic data'])\n", "('NONE', 'present a XXXXX based XXXXX', ['lagrangian relaxation', 'solution method'])\n", "('method used for task', 'a XXXXX is proposed to obtain good lower and XXXXX of the problem', ['hybrid approach', 'upper bounds'])\n", "('method used for task', 'we propose XXXXX to improve the XXXXX of the proposed formulation', ['valid inequalities', 'lower bound'])\n", "('NONE', 'for some XXXXX , the XXXXX is a simple threshold policy', ['optimal policy', 'parameter values'])\n", "('method used for task', 'we apply XXXXX and dynamic XXXXX to solve the problem', ['column generation', 'constraint aggregation'])\n", "('NONE', 'we improve lower and XXXXX on existing XXXXX', ['upper bounds', 'benchmark instances'])\n", "('method used for task', 'minimizing the real XXXXX is introduced as a practical XXXXX', ['travel time', 'robustness measure'])\n", "('method used for task', 'we propose XXXXX and XXXXX to model service times', ['finite mixtures', 'dirichlet processes'])\n", "('method used for task', 'we propose an innovative XXXXX for clustering categorical XXXXX', ['hybrid method', 'sequential data'])\n", "('NONE', 'we propose two XXXXX which determine the players’ XXXXX', ['optimal strategies', 'solution procedures'])\n", "('method used for task', 'this paper uses a XXXXX to measure the XXXXX index', ['probabilistic analysis', 'malmquist productivity'])\n", "('NONE', 'we propose a new XXXXX ; it computes the XXXXX for small cases', ['global optimum', 'milp formulation'])\n", "('method used for task', 'the scc XXXXX is modeled as a XXXXX ( mip ) problem', ['scheduling problem', 'mixed-integer programming'])\n", "('method used for task', 'we propose simple and XXXXX to solve those XXXXX', ['efficient algorithms', 'optimization problems'])\n", "('method used for task', 'a strongly XXXXX is proposed to solve the XXXXX', ['polynomial-time algorithm', 'inverse problem'])\n", "('method used for task', 'propose a XXXXX based on sequential bifurcation and XXXXX', ['response surface methodology', 'solution framework'])\n", "('NONE', 'we analyze an XXXXX ( ev ) XXXXX under a price-discount scheme', ['electric vehicle', 'supply chain'])\n", "('method used for task', 'use of XXXXX to improve XXXXX', ['social welfare', 'cross-border system'])\n", "('method used for task', 'a new XXXXX for XXXXX is proposed and used', ['classification scheme', 'transport operations'])\n", "('NONE', 'we provide explicit XXXXX on the multiple as function of past XXXXX and volatilities', ['upper bounds', 'asset returns'])\n", "('method used for task', 'we show how this new XXXXX performs for various financial XXXXX', ['asset allocation method', 'market conditions'])\n", "('method used for task', 'insufficiency of XXXXX to estimate an XXXXX is identified', ['linear relaxation solutions', 'upper bound set'])\n", "('method used for task', 'a heuristic using XXXXX and a XXXXX is proposed', ['priority rules', 'column generation algorithm'])\n", "('NONE', 'a XXXXX solves the problem with varying XXXXX', ['resource availability', 'column generation algorithm'])\n", "('NONE', 'we realise an XXXXX for hedge funds using XXXXX approach', ['early warning system', 'regression trees'])\n", "('method used for task', 'XXXXX and equal XXXXX are its boundary members', ['shapley value', 'division value'])\n", "('NONE', 'numerical applications in XXXXX of XXXXX', ['portfolio optimization', 'stock markets'])\n", "('NONE', 'we compute an XXXXX with a least XXXXX heuristic', ['upper bound', 'discrepancy search'])\n", "('NONE', 'we consider a two-echelon XXXXX with poisson demand and XXXXX', ['lost sales', 'inventory system'])\n", "('method used for task', 'constant XXXXX for XXXXX with sigmoid utility', ['knapsack problem', 'factor algorithm'])\n", "('method used for task', 'constant XXXXX for XXXXX with sigmoid utility', ['generalized assignment problem', 'factor algorithm'])\n", "('NONE', 'it initially deals with XXXXX in possibilistic XXXXX', ['dynamic systems', 'optimal stopping problems'])\n", "('method used for task', 'the presence of a unique reorder level wherever the XXXXX is insensitive to certain XXXXX', ['order quantity', 'parameter values'])\n", "('NONE', 'this simple XXXXX can be further refined making use of all the short-term XXXXX available', ['priority rule', 'demand information'])\n", "('method used for task', 'an extensive XXXXX is presented on node XXXXX with intermediate facilities', ['literature review', 'routing problems'])\n", "('method used for task', 'four XXXXX for amblp are analyzed for XXXXX', ['heuristic methods', 'worst-case performance'])\n", "('NONE', 'we model a flexible XXXXX with n market periods and XXXXX', ['demand uncertainty', 'capacity strategy'])\n", "('NONE', 'locating XXXXX by conventional methods makes XXXXX vulnerable', ['facts devices', 'power systems'])\n", "('NONE', 'give a model for the XXXXX of the legislative council of XXXXX ( legco )', ['voting system', 'hong kong'])\n", "('NONE', 'an XXXXX of u.s. coal-fired XXXXX operating in 2011', ['empirical study', 'power plants'])\n", "('method used for task', 'the portfolio problem solved using XXXXX and XXXXX', ['dynamic programming', 'monte carlo methods'])\n", "('NONE', 'the XXXXX solved using XXXXX and monte carlo methods', ['dynamic programming', 'portfolio problem'])\n", "('method used for task', 'the XXXXX solved using dynamic programming and XXXXX', ['monte carlo methods', 'portfolio problem'])\n", "('NONE', 'minimization of the XXXXX of any desired XXXXX of the hedging error', ['expected value', 'penalty function'])\n", "('method used for task', 'innovative integration of XXXXX and the well-known ( s , s ) XXXXX', ['capacity strategy', 'inventory policy'])\n", "('method used for task', 'an improved kfp-mco classifier based kernel , fuzzification , and XXXXX is proposed and used for XXXXX', ['credit scoring', 'penalty factors'])\n", "('method used for task', 'we illustrate that parametric bayesian updates based on observed XXXXX can be used effectively for XXXXX', ['sales data', 'demand learning'])\n", "('method used for task', 'dynamic XXXXX is larger before the XXXXX than during it', ['financial crisis', 'cost inefficiency'])\n", "('NONE', 'introducing a new XXXXX : tollbooth XXXXX', ['queueing model', 'tandem queue'])\n", "('method used for task', 'introducing some special XXXXX related to the tollbooth XXXXX', ['performance measures', 'tandem queue'])\n", "('NONE', 'developing the XXXXX and techniques that are the cornerstones of static XXXXX', ['network flows', 'key concepts'])\n", "('NONE', 'vehicle emissions and total XXXXX are considered in the upper XXXXX', ['travel time', 'level problem'])\n", "('NONE', 'XXXXX provision does have an impact on the number of XXXXX', ['traffic information', 'optimal designs'])\n", "('NONE', 'promoters and ribosome XXXXX categorized as low , medium and XXXXX', ['binding sites', 'high efficiency'])\n", "('method used for task', 'price-lead time-inventory trade-off with XXXXX and XXXXX', ['product variety', 'uncertain demand'])\n", "('NONE', 'study of the XXXXX of the XXXXX and the feasible region', ['geometric properties', 'level sets'])\n", "('method used for task', 'approaches with XXXXX and related XXXXX and costs are rare', ['multiple products', 'setup times'])\n", "('NONE', 'XXXXX , XXXXX', ['branch-and-price algorithm', 'column generation'])\n", "('method used for task', 'the expected cost of XXXXX is computed using a XXXXX', ['stochastic model', 'system operation'])\n", "('NONE', 'it is shown that XXXXX tracking information can yield significant XXXXX', ['real time', 'cost savings'])\n", "('NONE', 'we consider currency risk management in a XXXXX of a multinational firm with XXXXX and risk sharing contracts', ['supply chain', 'risk transfer'])\n", "('NONE', 'we consider currency risk management in a XXXXX of a multinational firm with risk transfer and XXXXX', ['supply chain', 'risk sharing contracts'])\n", "('method used for task', 'we consider currency risk management in a supply chain of a multinational firm with XXXXX and XXXXX', ['risk transfer', 'risk sharing contracts'])\n", "('NONE', 'we study and compare the impacts of XXXXX and risk sharing contracts on utilities of XXXXX members', ['supply chain', 'risk transfer'])\n", "('NONE', 'we study and compare the impacts of risk transfer and XXXXX on utilities of XXXXX members', ['supply chain', 'risk sharing contracts'])\n", "('method used for task', 'we study and compare the impacts of XXXXX and XXXXX on utilities of supply chain members', ['risk transfer', 'risk sharing contracts'])\n", "('NONE', 'the newly proposed models give reliable evaluation of XXXXX in XXXXX', ['system performance', 'heavy traffic'])\n", "('method used for task', 'a XXXXX for the XXXXX problem using conditional value at risk', ['decision support system', 'vehicle replacement'])\n", "('NONE', 'identification of the XXXXX in XXXXX', ['risk drivers', 'vehicle replacement'])\n", "('NONE', 'a faster minimum XXXXX ( msi ) XXXXX is presented', ['size instance', 'identification algorithm'])\n", "('NONE', 'we compute XXXXX using a semidefinite rather than a XXXXX', ['lower bounds', 'linear relaxation'])\n", "('NONE', 'it is the first paper that tries to regard XXXXX in reverse XXXXX', ['supply chain', 'risk measure'])\n", "('NONE', 'proposed heuristic score based subset XXXXX reduces the XXXXX into polynomial growth', ['search space', 'generation process'])\n", "('NONE', 'presence of in-city XXXXX has a positive impact on XXXXX', ['spare parts', 'maintenance financial performance'])\n", "('method used for task', 'developed XXXXX to calculate throughput and other XXXXX', ['markov chain model', 'performance indicators'])\n", "('NONE', 'used XXXXX to quantify impact of XXXXX on line performance', ['markov model', 'system parameters'])\n", "('NONE', 'used XXXXX to quantify impact of system parameters on XXXXX', ['markov model', 'line performance'])\n", "('NONE', 'used markov model to quantify impact of XXXXX on XXXXX', ['system parameters', 'line performance'])\n", "('method used for task', 'XXXXX is employed to examine the effects of XXXXX modification on the prioritization solution', ['sensitivity analysis', 'parameter values'])\n", "('method used for task', 'we formulate an adapted XXXXX model to select the XXXXX', ['optimal strategy', 'mean-variance portfolio'])\n", "('NONE', 'characterizing the XXXXX utilizing XXXXX', ['sensitivity analysis', 'anchor points'])\n", "('method used for task', 'a XXXXX and a XXXXX for the operational problem are developed', ['dynamic model', 'linear approximation'])\n", "('method used for task', 'a XXXXX discusses performance with respect to XXXXX', ['computational study', 'problem characteristics'])\n", "('NONE', 'computational results show very XXXXX of the proposed XXXXX', ['high efficiency', 'planning framework'])\n", "('NONE', 'the XXXXX increases not only sales but XXXXX and default risk', ['trade credit', 'opportunity cost'])\n", "('method used for task', 'the XXXXX increases not only sales but opportunity cost and XXXXX', ['trade credit', 'default risk'])\n", "('method used for task', 'the trade credit increases not only sales but XXXXX and XXXXX', ['opportunity cost', 'default risk'])\n", "('method used for task', 'both optimal XXXXX and optimal XXXXX not only exist but also are unique', ['trade credit', 'cycle time'])\n", "('NONE', 'we study XXXXX in serial multi-tier XXXXX', ['vertical integration', 'supply chains'])\n", "('NONE', 'multiobjective binary XXXXX under XXXXX is captured', ['linear programming', 'interval uncertainty'])\n", "('method used for task', 'incorporating XXXXX and co2 emission costs into an abc XXXXX', ['capacity expansion', 'decision model'])\n", "('method used for task', 'incorporating XXXXX and XXXXX costs into an abc decision model', ['capacity expansion', 'co2 emission'])\n", "('NONE', 'incorporating capacity expansion and XXXXX costs into an abc XXXXX', ['decision model', 'co2 emission'])\n", "('NONE', 'proposing a XXXXX in selecting green XXXXX', ['mathematical programming approach', 'building projects'])\n", "('NONE', 'management must incorporate XXXXX cost into overall green XXXXX costs', ['co2 emission', 'building project'])\n", "('NONE', 'we investigate the value of accounting for XXXXX in XXXXX', ['inventory control', 'demand seasonality'])\n", "('NONE', 'we compute the XXXXX of policies which ignore part or all XXXXX', ['optimality gaps', 'demand seasonality'])\n", "('method used for task', 'considering systemicity of risks ensures more XXXXX with respect to XXXXX', ['robust decision making', 'risk analysis'])\n", "('NONE', 'building risk maps using XXXXX approaches improves XXXXX', ['problem structuring', 'risk analysis'])\n", "('method used for task', 'new sampling algorithm giving XXXXX to XXXXX is proposed', ['fast convergence', 'uniform distribution'])\n", "('method used for task', 'a hybrid heuristic algorithm based on XXXXX and XXXXX are developed', ['simulated annealing', 'binary search'])\n", "('NONE', 'XXXXX of XXXXX ( mmds ) have three levels of representation : elements , loops and the whole model', ['mental models', 'dynamic systems'])\n", "('method used for task', 'an XXXXX and a XXXXX are developed', ['branch-and-cut algorithm', 'integer programming formulation'])\n", "('NONE', 'XXXXX from 318 XXXXX in the u.s', ['survey data', 'manufacturing firms'])\n", "('NONE', 'we develop a 4-approximation algorithm for incremental XXXXX with XXXXX', ['network design', 'shortest paths'])\n", "('method used for task', 'liner fleet XXXXX solved without requiring complete XXXXX', ['probability distributions', 'deployment problem'])\n", "('NONE', 'we have studied the mobile facility routing and XXXXX with XXXXX', ['scheduling problem', 'stochastic demand'])\n", "('method used for task', 'the reliability bounds provide an XXXXX for very XXXXX', ['approximate solution', 'large networks'])\n", "('method used for task', 'XXXXX of the digraph is very effective and greatly reduces XXXXX', ['parallel processing', 'computational time'])\n", "('NONE', 'we apply our model to XXXXX of XXXXX', ['breast cancer', 'radiation therapy treatment planning'])\n", "('NONE', 'the XXXXX improves all best XXXXX known in literature', ['lower bounds', 'lower bound methodology'])\n", "('NONE', 'a better heuristic XXXXX in finding minimal XXXXX is presented', ['solution technique', 'cost bounds'])\n", "('NONE', 'our algorithm , bbh , can solve real-world XXXXX with short XXXXX', ['computation time', 'size instances'])\n", "('NONE', 'bbh outperforms the current state of the art XXXXX for most of the XXXXX', ['solution methods', 'problem instances'])\n", "('NONE', 'the XXXXX of XXXXX is extended to the case of hrs model', ['malmquist index', 'productivity change'])\n", "('NONE', 'preference model is a set of XXXXX compatible with XXXXX', ['value functions', 'preference information'])\n", "('NONE', 'XXXXX is a set of XXXXX compatible with preference information', ['value functions', 'preference model'])\n", "('NONE', 'XXXXX is a set of value functions compatible with XXXXX', ['preference information', 'preference model'])\n", "('NONE', 'we find improved best XXXXX for 18 out of the 37 XXXXX', ['lower bounds', 'benchmark instances'])\n", "('method used for task', 'the technique sequentially minimizes risk in terms of XXXXX and XXXXX', ['waiting time', 'service time'])\n", "('method used for task', 'the technique applies to any convex XXXXX and XXXXX distribution', ['loss function', 'service time'])\n", "('NONE', 'the novel XXXXX concept of line-integrated supermarkets in order to supply XXXXX is introduced', ['assembly lines', 'part logistics'])\n", "('method used for task', 'we design a XXXXX for ship stowage XXXXX', ['classification scheme', 'planning problems'])\n", "('NONE', 'a XXXXX of XXXXX is established for symmetric unimodal demand distributions', ['lower bound', 'cost deviation'])\n", "('method used for task', 'the XXXXX is more sensitive than the XXXXX to sub-optimal ordering decisions', ['newsvendor model', 'eoq model'])\n", "('method used for task', 'a new distributed XXXXX is designed for approximating the XXXXX of bi-objective optimization problems ( dlbs )', ['heuristic approach', 'pareto set'])\n", "('method used for task', 'we develop mixed XXXXX and XXXXX for the proposed models', ['exact algorithms', 'integer programs'])\n", "('NONE', 'an estimation method is proposed for XXXXX in choice based XXXXX', ['value functions', 'conjoint analysis'])\n", "('method used for task', 'an XXXXX is proposed for XXXXX in choice based conjoint analysis', ['value functions', 'estimation method'])\n", "('method used for task', 'an XXXXX is proposed for value functions in choice based XXXXX', ['conjoint analysis', 'estimation method'])\n", "('NONE', 'we consider the role of XXXXX in XXXXX in decentralized assembly networks', ['knowledge sharing', 'cost reduction'])\n", "('NONE', 'application to truncation and XXXXX in XXXXX', ['inventory models', 'supply chain management'])\n", "('NONE', 'we develop a new XXXXX , hesitant XXXXX', ['optimization method', 'fuzzy programming method'])\n", "('method used for task', 'we compare 12 heuristics and 9 XXXXX for the simple XXXXX type 1', ['lower bounds', 'assembly line balancing problem'])\n", "('method used for task', 'the success of heuristics and XXXXX strongly depends on the XXXXX', ['lower bounds', 'problem characteristics'])\n", "('method used for task', 'we employ XXXXX and XXXXX using preference for resources', ['genetic algorithm', 'combinatorial auction'])\n", "('NONE', 'we analyze how the XXXXX affects the strategies in a two-tiers XXXXX', ['spot market', 'supply chain'])\n", "('NONE', 'we study how XXXXX affect the behavior and dynamic of the XXXXX members', ['risk factors', 'supply chain'])\n", "('NONE', 'we consider a XXXXX with one global constraint on XXXXX', ['carbon emissions', 'lot-sizing problem'])\n", "('method used for task', 'both optimal XXXXX and XXXXX not only exist but also are unique', ['trade credit', 'cycle time'])\n", "('NONE', 'XXXXX of XXXXX with perturbations on weights', ['robustness analysis', 'pareto optimal solutions'])\n", "('method used for task', 'new XXXXX ( svr ) techniques are applied to XXXXX of corporate bonds', ['support vector regression', 'recovery rates'])\n", "('NONE', 'transformation of XXXXX does not improve the XXXXX', ['prediction accuracy', 'recovery rates'])\n", "('method used for task', 'a heuristic XXXXX is implemented and compared to the XXXXX', ['greedy algorithm', 'exact method'])\n", "('NONE', 'XXXXX are obtained by solving single-objective XXXXX using cplex', ['pareto solutions', 'milp models'])\n", "('method used for task', 'a XXXXX is concerned with both efficient investment and XXXXX', ['supply chain', 'production decisions'])\n", "('NONE', 'the dynamic XXXXX utilizes multiple XXXXX including fim™ instrument , bbs and moca©/mmse', ['dea model', 'assessment criteria'])\n", "('method used for task', 'we study the XXXXX for two demand classes with heterogeneous XXXXX', ['order acceptance', 'lead times'])\n", "('method used for task', 'we study the XXXXX for two XXXXX with heterogeneous lead times', ['order acceptance', 'demand classes'])\n", "('NONE', 'we study the order acceptance for two XXXXX with heterogeneous XXXXX', ['lead times', 'demand classes'])\n", "('NONE', 'XXXXX , XXXXX , reversed XXXXX and stochastic ordering comparisons are carried out', ['likelihood ratio', 'failure rate'])\n", "('NONE', 'XXXXX , XXXXX , reversed XXXXX and stochastic ordering comparisons are carried out', ['likelihood ratio', 'failure rate'])\n", "('method used for task', 'we find a XXXXX where decentralization is better than XXXXX', ['time interval', 'vertical integration'])\n", "('NONE', 'deterministic formulation where XXXXX follow a XXXXX', ['uncertain parameters', 'normal distribution'])\n", "('method used for task', 'XXXXX and an efficient XXXXX', ['branch-and-cut algorithm', 'heuristic algorithm'])\n", "('NONE', 'a XXXXX should match with manufacturers’ XXXXX ( mc )', ['consignment contract', 'sales promotion'])\n", "('NONE', 'a wholesale XXXXX should match with retailers’ XXXXX ( rp )', ['price contract', 'sales promotion'])\n", "('method used for task', 'we combine the advantages of XXXXX and XXXXX', ['evolutionary algorithms', 'statistical model'])\n", "('method used for task', 'new access charges XXXXX on the XXXXX', ['model based', 'system approach'])\n", "('method used for task', 'it is an XXXXX that does not guarantee to catch the XXXXX', ['approximation method', 'global optimum'])\n", "('method used for task', 'we examine the impact of consumer environmental awareness ( cea ) on XXXXX and XXXXX', ['channel coordination', 'order quantities'])\n", "('NONE', 'a XXXXX can effectively coordinate a decentralized XXXXX', ['return policy', 'supply chain'])\n", "('NONE', 'we study the inconsistency of XXXXX in XXXXX', ['pairwise comparisons', 'group decision making'])\n", "('NONE', 'it benefits of the XXXXX , that can be solved with modest XXXXX', ['linear relaxation', 'computational time'])\n", "('method used for task', 'proof of existence of XXXXX for stochastic linear XXXXX', ['optimal policies', 'control problems'])\n", "('method used for task', 'the method adopts the concept of lca and integrates fuzzy XXXXX and XXXXX', ['fuzzy topsis', 'extent analysis'])\n", "('NONE', 'this study considers homebuyers’ XXXXX with s-shaped XXXXX', ['risk attitude', 'utility functions'])\n", "('NONE', 'we propose a cyclical square-root continuous-time model for the XXXXX of XXXXX', ['term structure', 'interest rates'])\n", "('method used for task', 'we use XXXXX and XXXXX for the pricing subproblem', ['benders’ decomposition', 'constraint programming'])\n", "('NONE', 'the role of XXXXX on XXXXX and productivity is assessed', ['social capital', 'production efficiency'])\n", "('NONE', 'introduce a method to measure XXXXX in XXXXX', ['operational risk', 'emerging markets'])\n", "('method used for task', 'XXXXX is higher in XXXXX than developed ones', ['operational risk', 'emerging markets'])\n", "('method used for task', 'XXXXX more dominant influence on XXXXX than sector', ['operational risk', 'market development'])\n", "('NONE', 'assessment of sensor reading weights based on XXXXX using the XXXXX', ['multiple criteria', 'ahp method'])\n", "('method used for task', 'XXXXX for XXXXX and spot prices including their interrelations', ['simulation model', 'wind power'])\n", "('method used for task', 'XXXXX for wind power and XXXXX including their interrelations', ['simulation model', 'spot prices'])\n", "('method used for task', 'simulation model for XXXXX and XXXXX including their interrelations', ['wind power', 'spot prices'])\n", "('NONE', 'the XXXXX of XXXXX is strongly influenced by the dependence structure', ['market value', 'wind turbines'])\n", "('NONE', 'the XXXXX of wind turbines is strongly influenced by the XXXXX', ['market value', 'dependence structure'])\n", "('method used for task', 'the market value of XXXXX is strongly influenced by the XXXXX', ['wind turbines', 'dependence structure'])\n", "('NONE', 'provides suggested routes for XXXXX that can combine key themes in analytics and XXXXX', ['future research', 'operational research'])\n", "('NONE', 'estimating XXXXX from inspections performed at irregular XXXXX', ['dirichlet distribution', 'time intervals'])\n", "('method used for task', 'the close relationship between XXXXX and XXXXX is studied', ['aggregation functions', 'spread measures'])\n", "('NONE', 'among spread measures we find many well-known data analysis tools like the XXXXX , XXXXX , mad , range , and iqr', ['standard deviation', 'sample variance'])\n", "('NONE', 'non-symmetric XXXXX may be used e.g . in XXXXX', ['group decision making', 'spread measures'])\n", "('NONE', 'we consider the impact of various aspects which may affect the estimation of probability of XXXXX in the context of XXXXX , such as uncertainty level , alternative preference scales and different weight estimation methods . we also consider the case where the comparisons are carried out in a fuzzy manner . it is revealed that the results hold regardless the aforementioned aspects considered', ['rank reversal', 'pairwise comparisons'])\n", "('NONE', 'we consider the impact of various aspects which may affect the estimation of probability of XXXXX in the context of pairwise comparisons , such as XXXXX , alternative preference scales and different weight estimation methods . we also consider the case where the comparisons are carried out in a fuzzy manner . it is revealed that the results hold regardless the aforementioned aspects considered', ['rank reversal', 'uncertainty level'])\n", "('NONE', 'we consider the impact of various aspects which may affect the estimation of probability of rank reversal in the context of XXXXX , such as XXXXX , alternative preference scales and different weight estimation methods . we also consider the case where the comparisons are carried out in a fuzzy manner . it is revealed that the results hold regardless the aforementioned aspects considered', ['pairwise comparisons', 'uncertainty level'])\n", "('method used for task', 'we use convex XXXXX to combine scheduling and XXXXX', ['non-linear programming', 'capacity planning'])\n", "('method used for task', 'we provide an XXXXX that captures the key trade-offs among time-to-market , XXXXX , pricing and production decisions', ['optimization model', 'sales channel'])\n", "('method used for task', 'we provide an XXXXX that captures the key trade-offs among time-to-market , sales channel , pricing and XXXXX', ['optimization model', 'production decisions'])\n", "('method used for task', 'we provide an optimization model that captures the key trade-offs among time-to-market , XXXXX , pricing and XXXXX', ['sales channel', 'production decisions'])\n", "('NONE', 'we characterize XXXXX , XXXXX and production policies', ['optimal pricing', 'market timing'])\n", "('NONE', 'we study the role of the secondary XXXXX on XXXXX , pricing and inventory decisions', ['sales channel', 'market timing'])\n", "('method used for task', 'we study the role of the secondary XXXXX on market timing , pricing and XXXXX', ['sales channel', 'inventory decisions'])\n", "('method used for task', 'we study the role of the secondary sales channel on XXXXX , pricing and XXXXX', ['market timing', 'inventory decisions'])\n", "('method used for task', 'we provide a method to characterize XXXXX for XXXXX', ['optimal policies', 'optimal stopping problems'])\n", "('NONE', \"we model a product 's degradation path by a wiener stochastic process and investigate the XXXXX of a step-stress accelerated degradation test ( ssadt ) for obtaining precise estimates of XXXXX\", ['optimum design', 'model parameters'])\n", "('NONE', 'changes in XXXXX of discrete XXXXX over time investigated', ['survival models', 'parameter estimates'])\n", "('method used for task', 'an improved h-mk-svm is developed to integrate the external , tag and keyword , individual behavioral and engagement behavioral data for XXXXX from multiple correlated attributes and XXXXX', ['feature selection', 'ensemble learning'])\n", "('NONE', 'consider uncertainty and XXXXX in XXXXX first time', ['multiple objectives', 'traveling salesman problem'])\n", "('NONE', 'we analyze a situation in which several firms deal with XXXXX concerning the same type of product and having XXXXX warehouses', ['limited capacity', 'inventory problems'])\n", "('NONE', 'we model a period-review XXXXX with XXXXX and fixed order costs', ['lost sales', 'inventory system'])\n", "('NONE', 'XXXXX considers the impact of XXXXX on financial efficiency', ['dynamic model', 'environmental variables'])\n", "('NONE', 'by focusing on linear and affine XXXXX we justify the use of new XXXXX', ['random variables', 'uncertainty sets'])\n", "('method used for task', 'the zone based XXXXX is modeled as a XXXXX', ['vehicle routing problem', 'multi-objective problem'])\n", "('method used for task', 'XXXXX asymmetry is confined to developed markets during XXXXX downtowns only', ['stock market', 'return reversal'])\n", "('method used for task', 'significantly improve best-known XXXXX for XXXXX', ['solution values', 'benchmark problem instances'])\n", "('method used for task', 'working environment affects operators’ XXXXX and the XXXXX', ['system performance', 'health condition'])\n", "('method used for task', 'work-related ill health ( wih ) XXXXX are considered to model XXXXX of the operators', ['risk factors', 'health states'])\n", "('method used for task', 'we develop efficient XXXXX based on XXXXX , vlsn search , and a hybrid algorithm that integrates the two', ['heuristic algorithms', 'tabu search'])\n", "('method used for task', 'we develop efficient XXXXX based on tabu search , vlsn search , and a XXXXX that integrates the two', ['heuristic algorithms', 'hybrid algorithm'])\n", "('method used for task', 'we develop efficient heuristic algorithms based on XXXXX , vlsn search , and a XXXXX that integrates the two', ['tabu search', 'hybrid algorithm'])\n", "('NONE', 'the XXXXX establishes that effective integration of simple XXXXX with vlsn search results in superior outcomes', ['computational study', 'tabu search'])\n", "('method used for task', 'we propose a dynamic pricing model with XXXXX with different behaviors applied to the XXXXX', ['multiple products', 'airline industry'])\n", "('NONE', 'we propose a XXXXX with XXXXX with different behaviors applied to the airline industry', ['multiple products', 'dynamic pricing model'])\n", "('method used for task', 'we propose a XXXXX with multiple products with different behaviors applied to the XXXXX', ['airline industry', 'dynamic pricing model'])\n", "('NONE', 'XXXXX eats green product’s sales more under certain XXXXX', ['limited capacity', 'market conditions'])\n", "('method used for task', 'a modified silver–meal heuristic combined with a XXXXX is proposed to overcome XXXXX', ['safety stock', 'demand variability'])\n", "('NONE', 'we study XXXXX policies under which consumers’ valuation depends on both the refund amount and the length of the XXXXX', ['consumer returns', 'return deadline'])\n", "('method used for task', 'we put forward a new differentiated buy-back contract contingent on the XXXXX to coordinate the XXXXX', ['supply chain', 'return deadline'])\n", "('NONE', 'new type of indirect XXXXX in form of assignment-based XXXXX of alternatives', ['preference information', 'pairwise comparisons'])\n", "('NONE', 'applications on XXXXX as well as on two XXXXX show that the proposed approaches outperform traditional techniques for conjoint analysis', ['experimental data', 'empirical studies'])\n", "('method used for task', 'applications on XXXXX as well as on two empirical studies show that the proposed approaches outperform traditional techniques for XXXXX', ['experimental data', 'conjoint analysis'])\n", "('method used for task', 'applications on experimental data as well as on two XXXXX show that the proposed approaches outperform traditional techniques for XXXXX', ['empirical studies', 'conjoint analysis'])\n", "('NONE', 'validating the proposed model and XXXXX via a real XXXXX', ['case study', 'solution technique'])\n", "('NONE', 'we design traditional XXXXX using different XXXXX', ['classification methods', 'failure models'])\n", "('method used for task', 'we propose a new cost-efficiency measure based on the notions of ray XXXXX and optimal XXXXX', ['average cost', 'scale size'])\n", "('method used for task', \"the new XXXXX ( ace ) extends banker and thrall 's ( 1992 ) tsre measure to XXXXX\", ['cost analysis', 'efficiency measure'])\n", "('method used for task', 'we show new lower and XXXXX on the XXXXX of spt', ['upper bounds', 'worst-case ratio'])\n", "('NONE', 'we structure the field of XXXXX in the XXXXX', ['automotive industry', 'part logistics'])\n", "('method used for task', 'XXXXX based on XXXXX is a latest problem in d–s theory', ['parameter estimation', 'belief structure'])\n", "('method used for task', 'XXXXX based on interval-valued XXXXX is first studied', ['parameter estimation', 'belief structure'])\n", "('method used for task', 'an incremental XXXXX is studied , where the objective is to maximize the cumulative s-t-flow over a XXXXX', ['network design problem', 'time horizon'])\n", "('NONE', 'the XXXXX as well as the heuristics are compared in XXXXX on randomly generated instances', ['computational experiments', 'mip formulations'])\n", "('NONE', 'we develop an minlp model which determines the XXXXX of a mixed ac–dc building electrical XXXXX', ['optimal design', 'distribution system'])\n", "('NONE', 'the XXXXX with simultaneous pickup and delivery with XXXXX is presented and the literature is summarized', ['vehicle routing problem', 'time limit'])\n", "('NONE', 'a number of well-known XXXXX with XXXXX and no XXXXX are solved and the solutions are compared', ['benchmark problems', 'time limit'])\n", "('NONE', 'a number of well-known XXXXX with XXXXX and no XXXXX are solved and the solutions are compared', ['benchmark problems', 'time limit'])\n", "('method used for task', 'a perturbation based variable neighborhood search heuristic proposed new best and XXXXX for XXXXX', ['robust solutions', 'benchmark problem instances'])\n", "('NONE', 'employing the XXXXX in XXXXX without taking into account that the distribution is mixed normal leads to a loss of 2–3 percent per year', ['portfolio selection', 'mean-variance rule'])\n", "('NONE', 'the XXXXX is determined from the variability of each indicator projected onto XXXXX', ['principal components', 'preference structure'])\n", "('method used for task', 'nonlinear XXXXX are suitable for agricultural XXXXX', ['policy analysis', 'mean-variance models'])\n", "('method used for task', 'a two-stage XXXXX with an order-up-to XXXXX is considered', ['supply chain', 'inventory policy'])\n", "('NONE', 'two new procedures using a multi-attribute XXXXX with incomplete XXXXX on weights related to performance measures', ['utility function', 'preference information'])\n", "('method used for task', 'two new procedures using a multi-attribute XXXXX with incomplete preference information on weights related to XXXXX', ['utility function', 'performance measures'])\n", "('method used for task', 'two new procedures using a multi-attribute utility function with incomplete XXXXX on weights related to XXXXX', ['preference information', 'performance measures'])\n", "('NONE', 'a XXXXX of risk propagation in a XXXXX is developed', ['supply network', 'bayesian network model'])\n", "('NONE', 'XXXXX demonstrates how the measures are used in a XXXXX setting', ['simulation study', 'supply network'])\n", "('NONE', 'we propose an XXXXX of a two-echelon recycled pulp XXXXX', ['agent-based model', 'supply chain'])\n", "('method used for task', 'XXXXX for u-shaped XXXXX are investigated', ['optimization models', 'assembly lines'])\n", "('NONE', 'we test the benefits of XXXXX , optimizing the periodic review interval , versus XXXXX , changing to a continuous review policy', ['parameter optimization', 'policy improvement'])\n", "('NONE', 'XXXXX outperforms XXXXX in most instances tested', ['parameter optimization', 'policy improvement'])\n", "('NONE', 'XXXXX for solutions of multi objective XXXXX are proposed', ['ranking methods', 'optimization problems'])\n", "('method used for task', 'we propose an XXXXX for the minimum common string XXXXX', ['ilp model', 'partition problem'])\n", "('method used for task', 'semi-open jackson XXXXX for XXXXX', ['network model', 'problem formulation'])\n", "('method used for task', 'XXXXX to find the XXXXX', ['iterative algorithm', 'optimal threshold'])\n", "('method used for task', 'the modelization provides very good XXXXX especially for rank 2 XXXXX matrix', ['lp relaxation', 'separation time'])\n", "('method used for task', 'XXXXX leads to a XXXXX based on constraint generation', ['best approximation', 'heuristic method'])\n", "('method used for task', 'XXXXX leads to a heuristic method based on XXXXX', ['best approximation', 'constraint generation'])\n", "('method used for task', 'best approximation leads to a XXXXX based on XXXXX', ['heuristic method', 'constraint generation'])\n", "('NONE', 'in context , XXXXX of several problems of XXXXX is analyzed', ['computational complexity', 'simple games'])\n", "('NONE', 'several XXXXX with XXXXX are presented sequentially over time', ['multiple attributes', 'decision alternatives'])\n", "('NONE', 'we extend the use of the combined stochastic dea–bayesian model to examine the XXXXX of XXXXX generated through the metafrontier', ['statistical properties', 'efficiency scores'])\n", "('method used for task', 'first a XXXXX on the XXXXX is shown ; then an exponential-time online optimal algorithm for the problem is presented ; and lastly a polynomial-time online asymptotically optimal algorithm is presented', ['lower bound', 'competitive ratio'])\n", "('method used for task', 'first a XXXXX on the competitive ratio is shown ; then an exponential-time online XXXXX for the problem is presented ; and lastly a polynomial-time online asymptotically XXXXX is presented', ['lower bound', 'optimal algorithm'])\n", "('method used for task', 'first a lower bound on the XXXXX is shown ; then an exponential-time online XXXXX for the problem is presented ; and lastly a polynomial-time online asymptotically XXXXX is presented', ['competitive ratio', 'optimal algorithm'])\n", "('NONE', 'experimental results show that the polynomial-time XXXXX takes only a fraction of the XXXXX of the offline optimal algorithm , yet produces solutions of competitive quality to the offline optimal solutions', ['online algorithm', 'running time'])\n", "('method used for task', 'we combine XXXXX and XXXXX approaches', ['stochastic programming', 'stochastic optimal control'])\n", "('NONE', 'we generate XXXXX with XXXXX and events of death', ['scenario trees', 'asset returns'])\n", "('NONE', 'the choice of the XXXXX at the customization point affects the total XXXXX', ['service level', 'operating cost'])\n", "('NONE', 'the choice of the XXXXX at the XXXXX affects the total operating cost', ['service level', 'customization point'])\n", "('NONE', 'the choice of the service level at the XXXXX affects the total XXXXX', ['operating cost', 'customization point'])\n", "('NONE', 'a non-constant XXXXX impacts the XXXXX', ['service level', 'customization point'])\n", "('NONE', 'focusing on determining the XXXXX with minimum total XXXXX in a dsm', ['activity sequence', 'feedback time'])\n", "('NONE', 'we solve two routing and XXXXX with split pickup and XXXXX', ['scheduling problems', 'split delivery'])\n", "('NONE', 'the discrete XXXXX provides quicker results but less XXXXX', ['split model', 'solution quality'])\n", "('method used for task', 'we provide an XXXXX to minimize the total delay cost , yielding XXXXX in short time', ['exact algorithm', 'exact solutions'])\n", "('method used for task', 'novel adaptation of efficient XXXXX to XXXXX', ['global optimization', 'dynamic environments'])\n", "('method used for task', 'the sensitivity to different XXXXX and XXXXX is studied', ['model parameters', 'cost parameters'])\n", "('NONE', 'XXXXX on response and recovery planning phases of XXXXX', ['literature survey', 'disaster management'])\n", "('NONE', 'for the XXXXX with a buffer , an XXXXX is developed for solving it', ['scheduling problem', 'optimal algorithm'])\n", "('method used for task', 'we derive XXXXX to solve the problems in XXXXX', ['sufficient conditions', 'polynomial time'])\n", "('method used for task', 'two metaheuristics are proposed based on XXXXX and XXXXX', ['neighborhood search', 'tree search'])\n", "('method used for task', 'we consider identical XXXXX to minimize XXXXX', ['parallel machine scheduling', 'total tardiness'])\n", "('method used for task', 'XXXXX and a XXXXX solve small problems', ['mathematical programming', 'branch-and-bound algorithm'])\n", "('method used for task', 'a XXXXX and a XXXXX are developed for larger problems', ['tabu search', 'hybrid genetic algorithm'])\n", "('NONE', 'XXXXX in XXXXX can lead to lower order rates', ['information sharing', 'supply chains'])\n", "('method used for task', 'a capital XXXXX for XXXXX has been suggested', ['coherent risk measures', 'allocation scheme'])\n", "('method used for task', 'focus on lagrangian decomposition , XXXXX , and XXXXX', ['column generation', 'benders’ decomposition'])\n", "('NONE', 'XXXXX increases not only sales but also XXXXX and default risk', ['trade credit', 'opportunity cost'])\n", "('method used for task', 'XXXXX increases not only sales but also opportunity cost and XXXXX', ['trade credit', 'default risk'])\n", "('method used for task', 'trade credit increases not only sales but also XXXXX and XXXXX', ['opportunity cost', 'default risk'])\n", "('method used for task', 'we use XXXXX to XXXXX in prices and network capacity', ['stochastic programming', 'model uncertainty'])\n", "('method used for task', 'we use XXXXX to model uncertainty in prices and XXXXX', ['stochastic programming', 'network capacity'])\n", "('NONE', 'we use stochastic programming to XXXXX in prices and XXXXX', ['model uncertainty', 'network capacity'])\n", "('NONE', 'XXXXX on the loss from delaying exercise of XXXXX are found', ['upper bounds', 'american options'])\n", "('NONE', 'we develop a novel XXXXX of XXXXX', ['probabilistic model', 'audit quality'])\n", "('NONE', 'we find that firms often adopt a XXXXX a XXXXX after its introduction', ['new technology', 'time lag'])\n", "('NONE', 'we study stochastic XXXXX , a class of stochastic XXXXX', ['cooperative games', 'linear programming games'])\n", "('NONE', 'for some certainty equivalents , a XXXXX can be computed in XXXXX', ['polynomial time', 'core allocation'])\n", "('NONE', 'approach a new pickup and XXXXX with lifo and XXXXX', ['time constraints', 'delivery problem'])\n", "('NONE', 'evaluate the performance of the proposed XXXXX through a comprehensive XXXXX', ['computational study', 'solution methods'])\n", "('method used for task', 'we propose a new XXXXX model for XXXXX', ['option pricing', 'random volatility'])\n", "('method used for task', 'an XXXXX is developed to solve the polynomial XXXXX', ['efficient algorithm', 'optimization problem'])\n", "('NONE', 'output XXXXX causes ex post XXXXX', ['price uncertainty', 'profit loss'])\n", "('NONE', 'XXXXX and technical inefficiency are the main sources of XXXXX', ['risk preference', 'profit loss'])\n", "('method used for task', 'we also consider a XXXXX and provide XXXXX on the number of pmus', ['theoretical model', 'lower bounds'])\n", "('NONE', 'the algorithm can achieve the optimum in the XXXXX of the XXXXX', ['theoretical model', 'power systems'])\n", "('NONE', 'we developed an XXXXX that solves the XXXXX', ['exact algorithm', 'feature selection problem'])\n", "('method used for task', 'considerable improvements in XXXXX and XXXXX over earlier methods', ['running time', 'solution quality'])\n", "('NONE', 'g-anp is specially useful for XXXXX with XXXXX', ['big data', 'decision making problems'])\n", "('NONE', 'it is applicable to the production of XXXXX in a XXXXX', ['multiple products', 'job shop'])\n", "('NONE', 'introduction of XXXXX extending the notion of XXXXX', ['aggregation functions', 'fusion functions'])\n", "('NONE', 'we provide a time window XXXXX in a vrp setting with XXXXX', ['demand uncertainty', 'assignment model'])\n", "('NONE', 'XXXXX show that using multiple XXXXX is better than one', ['numerical experiments', 'demand scenarios'])\n", "('method used for task', 'they are applicable to any number of XXXXX and any XXXXX', ['demand classes', 'demand distribution'])\n", "('method used for task', 'we consider sustainability issues on the joint XXXXX and XXXXX', ['trade credit', 'inventory decisions'])\n", "('NONE', 'XXXXX can improve consumer confidence and XXXXX', ['traceability systems', 'supply chain profit'])\n", "('NONE', 'we analyze the impact of XXXXX on XXXXX', ['failure interaction', 'warranty cost'])\n", "('method used for task', 'we carry out a real XXXXX to illustrate the XXXXX proposed', ['case study', 'multicriteria model'])\n", "('method used for task', 'we define an XXXXX for the bi-objective bi-dimensional XXXXX', ['knapsack problem', 'upper bound set'])\n", "('method used for task', 'XXXXX is relevant to tactical multi-project XXXXX', ['capacity planning', 'resource loading'])\n", "('NONE', 'to study a book XXXXX through XXXXX', ['supply chain', 'mathematical modeling'])\n", "('method used for task', 'the impact of XXXXX and different XXXXX is investigated', ['load distribution', 'objective functions'])\n", "('method used for task', 'the scope of or practice has been extended via XXXXX ( soft or ) and XXXXX', ['problem structuring methods', 'business analytics'])\n", "('NONE', 'to address how to improve the XXXXX in a two-level decentralized XXXXX with random yield and uncertain demand', ['service level', 'supply chain'])\n", "('NONE', 'to address how to improve the XXXXX in a two-level decentralized supply chain with XXXXX and uncertain demand', ['service level', 'random yield'])\n", "('NONE', 'to address how to improve the XXXXX in a two-level decentralized supply chain with random yield and XXXXX', ['service level', 'uncertain demand'])\n", "('NONE', 'to address how to improve the service level in a two-level decentralized XXXXX with XXXXX and uncertain demand', ['supply chain', 'random yield'])\n", "('NONE', 'to address how to improve the service level in a two-level decentralized XXXXX with random yield and XXXXX', ['supply chain', 'uncertain demand'])\n", "('method used for task', 'to address how to improve the service level in a two-level decentralized supply chain with XXXXX and XXXXX', ['random yield', 'uncertain demand'])\n", "('NONE', 'we provide the XXXXX of XXXXX for all the problems', ['lower bounds', 'competitive ratio'])\n", "('method used for task', 'development of various XXXXX and XXXXX', ['lower bounds', 'dominance rules'])\n", "('method used for task', 'a XXXXX is formulated to minimize the total XXXXX', ['operating cost', 'dynamic programming model'])\n", "('method used for task', 'XXXXX and a XXXXX are developed', ['branch-and-bound algorithm', 'milp formulations'])\n", "('method used for task', 'XXXXX is currently the best XXXXX', ['branch-and-bound algorithm', 'exact method'])\n", "('method used for task', 'we develop a pseudo-polynomial XXXXX for concave XXXXX', ['time algorithm', 'compression cost function'])\n", "('method used for task', 'we develop a XXXXX for convex XXXXX', ['polynomial time algorithm', 'compression cost function'])\n", "('method used for task', 'we develop a new XXXXX for minmax regret XXXXX', ['lower bound', 'combinatorial optimization problems'])\n", "('method used for task', 'our solution reduces two costs linked to this problem : XXXXX and handling XXXXX', ['fuel consumption', 'operations costs'])\n", "('NONE', 'the XXXXX can be found by a XXXXX', ['dynamic programming algorithm', 'optimal routing strategy'])\n", "('NONE', 'the XXXXX has a specific XXXXX', ['optimal strategy', 'threshold structure'])\n", "('NONE', 'XXXXX and the value of the marginal product of inputs are measured in one step using various XXXXX', ['technical efficiency', 'model specifications'])\n", "('NONE', 'we use a bootstrap dea technique to estimate the mean and 95 percent XXXXX of XXXXX and shadow prices', ['confidence intervals', 'technical efficiency'])\n", "('NONE', 'we use a bootstrap dea technique to estimate the mean and 95 percent XXXXX of technical efficiency and XXXXX', ['confidence intervals', 'shadow prices'])\n", "('method used for task', 'we use a bootstrap dea technique to estimate the mean and 95 percent confidence intervals of XXXXX and XXXXX', ['technical efficiency', 'shadow prices'])\n", "('NONE', 'we show how to approximate the XXXXX via XXXXX', ['efficient frontier', 'integer programming'])\n", "('method used for task', 'XXXXX to optimise biomass-based XXXXX', ['mathematical model', 'supply chains'])\n", "('NONE', 'an XXXXX with XXXXX where positive lead times are allowed', ['inventory model', 'random yield'])\n", "('NONE', 'an XXXXX with random yield where positive XXXXX are allowed', ['inventory model', 'lead times'])\n", "('NONE', 'an inventory model with XXXXX where positive XXXXX are allowed', ['random yield', 'lead times'])\n", "('NONE', 'we introduce an XXXXX of optimal value and XXXXX', ['upper bound', 'heuristic algorithms'])\n", "('method used for task', 'we state new production axioms that account for XXXXX and formally derive XXXXX from them', ['ratio data', 'dea models'])\n", "('method used for task', 'establish that expected loss in general stochastic XXXXX is non-increasing as XXXXX measured using entropy is reduced', ['demand uncertainty', 'inventory problems'])\n", "('NONE', 'classified by , e.g. , XXXXX , input data , objective , XXXXX', ['model formulation', 'solution method'])\n", "('method used for task', 'we present mean-variance models for hybrid XXXXX and translate them into XXXXX', ['portfolio optimization', 'convex quadratic programming'])\n", "('method used for task', 'we consider the XXXXX and give the XXXXX in the case with no more than two new securities', ['analytical solutions', 'solution procedures'])\n", "('method used for task', 'empirical estimates indicate difference in the parameter coefficients of gamma stochastic XXXXX , and heterogeneity function variables between the pooled and the swamy–arora XXXXX', ['production function', 'panel estimator'])\n", "('NONE', 'consider the XXXXX with XXXXX , multi-factor volatility and jumps', ['mean reversion', 'asset dynamics'])\n", "('NONE', 'we derive XXXXX to various types of discrete XXXXX', ['closed-form solutions', 'variance swaps'])\n", "('NONE', 'the XXXXX with a small XXXXX rate prefers the pooling strategy', ['service provider', 'demand loss'])\n", "('NONE', 'the XXXXX with a small demand loss rate prefers the XXXXX', ['service provider', 'pooling strategy'])\n", "('NONE', 'the service provider with a small XXXXX rate prefers the XXXXX', ['demand loss', 'pooling strategy'])\n", "('NONE', 'we describe the search region in XXXXX by local XXXXX', ['multi-objective optimization', 'upper bounds'])\n", "('NONE', 'we describe the XXXXX in XXXXX by local upper bounds', ['multi-objective optimization', 'search region'])\n", "('NONE', 'we describe the XXXXX in multi-objective optimization by local XXXXX', ['upper bounds', 'search region'])\n", "('NONE', 'we reveal XXXXX of the optimal instalment payment , which is different from that under XXXXX', ['structural properties', 'moral hazard'])\n", "('NONE', 'we consider a XXXXX with a piecewise linear XXXXX', ['transportation problem', 'cost structure'])\n", "('method used for task', 'we propose a XXXXX for stowage XXXXX of a container ship', ['heuristic algorithm', 'planning problem'])\n", "('method used for task', 'we give a framework for XXXXX and dynamic XXXXX', ['risk measures', 'stochastic optimization problems'])\n", "('method used for task', 'the XXXXX ( mopath ) is embedded in XXXXX', ['fuzzy model', 'optimization framework'])\n", "('method used for task', 'approximation based on XXXXX provides good solutions and XXXXX', ['network flows', 'lower bounds'])\n", "('NONE', 'XXXXX , dynamic , XXXXX , heterogeneous fleet constraints are handled', ['time windows', 'split delivery'])\n", "('method used for task', 'a XXXXX is also developed for larger instances which produces good XXXXX', ['heuristic procedure', 'approximate solutions'])\n", "('NONE', 'we propose a XXXXX that takes into account both cancellations and XXXXX behaviour', ['revenue management model', 'customer choice'])\n", "('method used for task', 'we develop an XXXXX for scheduling chp units with XXXXX', ['efficient algorithm', 'electric vehicles'])\n", "('method used for task', 'XXXXX for XXXXX with redials and reconnects ;', ['queueing model', 'call centers'])\n", "('NONE', 'we aim to maximise shareholder wealth while mitigating XXXXX of XXXXX in manufacturing', ['environmental impact', 'carbon emission'])\n", "('method used for task', 'studies in joint configuration of XXXXX and XXXXX are reviewed', ['supply chains', 'product families'])\n", "('method used for task', 'XXXXX for product families and XXXXX are formulated', ['mathematical models', 'supply chains'])\n", "('method used for task', 'XXXXX for XXXXX and supply chains are formulated', ['mathematical models', 'product families'])\n", "('method used for task', 'mathematical models for XXXXX and XXXXX are formulated', ['supply chains', 'product families'])\n", "('NONE', 'the socp approximations are used to obtain XXXXX at the top levels of a branch and XXXXX', ['lower bounds', 'bound algorithm'])\n", "('method used for task', 'we propose a three-stage XXXXX e-nautilus for computationally expensive XXXXX', ['interactive method', 'multiobjective optimization problems'])\n", "('NONE', 'a set of pre-calculated XXXXX enables a solution process without XXXXX', ['pareto optimal solutions', 'waiting times'])\n", "('NONE', 'a set of pre-calculated XXXXX enables a XXXXX without waiting times', ['pareto optimal solutions', 'solution process'])\n", "('NONE', 'a set of pre-calculated pareto optimal solutions enables a XXXXX without XXXXX', ['waiting times', 'solution process'])\n", "('method used for task', 'no new XXXXX is solved when the XXXXX interacts with the method', ['optimization problem', 'decision maker'])\n", "('method used for task', 'first result to comprehensively determine the XXXXX under endogenous and exogenous XXXXX', ['optimal policy', 'demand uncertainty'])\n", "('NONE', 'a XXXXX of published XXXXX for managing supply chain risks', ['systematic review', 'analytical models'])\n", "('NONE', 'a XXXXX of published analytical models for managing XXXXX', ['systematic review', 'supply chain risks'])\n", "('method used for task', 'a systematic review of published XXXXX for managing XXXXX', ['analytical models', 'supply chain risks'])\n", "('NONE', 'eight XXXXX for formal modeling of XXXXX', ['research streams', 'supply chain risks'])\n", "('method used for task', 'we propose a strong XXXXX ( mip ) for the stand XXXXX', ['mixed integer programming', 'allocation problem'])\n", "('NONE', 'it is shown that XXXXX can solve efficiently the XXXXX', ['greedy algorithms', 'view selection problem'])\n", "('NONE', 'the XXXXX perturbs a solution using XXXXX', ['simulated annealing', 'solution approach'])\n", "('method used for task', 'we present a XXXXX to solve resource-constrained XXXXX with flexible project structure', ['genetic algorithm', 'scheduling problems'])\n", "('method used for task', 'we present a XXXXX to solve resource-constrained scheduling problems with flexible XXXXX', ['genetic algorithm', 'project structure'])\n", "('NONE', 'we present a genetic algorithm to solve resource-constrained XXXXX with flexible XXXXX', ['scheduling problems', 'project structure'])\n", "('NONE', \"XXXXX of the shape of the firm 's XXXXX is presented\", ['empirical assessment', 'cost function'])\n", "('NONE', 'XXXXX are obtained under exponential XXXXX', ['closed-form solutions', 'utility function'])\n", "('NONE', 'we propose an approach for achieving XXXXX in XXXXX', ['customer satisfaction', 'manufacturing firms'])\n", "('NONE', 'a XXXXX was conducted in 24 XXXXX from 3 countries', ['questionnaire survey', 'manufacturing firms'])\n", "('NONE', 'the model is fitted to univariate inflow XXXXX by XXXXX', ['time series', 'quantile regression'])\n", "('method used for task', 'the XXXXX is to transfer all data in given time at the XXXXX', ['scheduling problem', 'minimum cost'])\n", "('NONE', 'advance notice , XXXXX , XXXXX and correlation are considered', ['lead times', 'random yield'])\n", "('NONE', 'longer remanufacturing XXXXX could reduce XXXXX', ['lead times', 'inventory variance'])\n", "('NONE', 'optimizing the removal of a XXXXX has low impact on XXXXX', ['base station', 'survival probability'])\n", "('NONE', 'we study the effect of information updating in a XXXXX with XXXXX', ['supply chain', 'spot market'])\n", "('method used for task', 'the new XXXXX also updates the belief on the XXXXX , and vice versa', ['demand information', 'spot price'])\n", "('method used for task', 'giving a XXXXX for obtaining the XXXXX in fdh models', ['polynomial-time algorithm', 'anchor points'])\n", "('NONE', 'find the role of XXXXX and time horizon in the probability of adoption of XXXXX', ['initial conditions', 'renewable energy'])\n", "('method used for task', 'find the role of XXXXX and XXXXX in the probability of adoption of renewable energy', ['initial conditions', 'time horizon'])\n", "('NONE', 'find the role of initial conditions and XXXXX in the probability of adoption of XXXXX', ['renewable energy', 'time horizon'])\n", "('NONE', 'trade resistance and XXXXX modeled as an undesirable XXXXX', ['production process', 'trade barriers'])\n", "('method used for task', 'a new XXXXX is also suggested for two XXXXX', ['calibration procedure', 'factor models'])\n", "('method used for task', 'we propose an XXXXX to optimize the lru XXXXX', ['milp model', 'definition decision problem'])\n", "('NONE', 'presenting XXXXX based on proven properties of the XXXXX', ['optimal algorithms', 'cost function'])\n", "('NONE', 'we tested the algorithm on XXXXX from the XXXXX', ['uci repository', 'benchmark instances'])\n", "('NONE', 'we compare XXXXX , XXXXX and deterministic approaches', ['robust optimization', 'stochastic programming'])\n", "('NONE', 'the XXXXX scales well with the size of the XXXXX', ['power system', 'robust optimization approach'])\n", "('NONE', 'models for the XXXXX of a district XXXXX were developed', ['optimal design', 'cooling system'])\n", "('NONE', 'XXXXX with XXXXX dependent cost function', ['parallel machine scheduling', 'completion time'])\n", "('NONE', 'XXXXX with completion time dependent XXXXX', ['parallel machine scheduling', 'cost function'])\n", "('NONE', 'parallel machine scheduling with XXXXX dependent XXXXX', ['completion time', 'cost function'])\n", "('method used for task', 'XXXXX solved via novel XXXXX', ['decomposition method', 'mixed integer programming model'])\n", "('method used for task', 'we develop a behavioral XXXXX for revenue sharing contracts incorporating XXXXX', ['decision model', 'reference points'])\n", "('method used for task', 'we conduct three XXXXX to test our XXXXX', ['laboratory experiments', 'behavioral model'])\n", "('method used for task', 'we model the problem with XXXXX and XXXXX', ['mixed integer programming', 'constraint programming'])\n", "('method used for task', 'bfp has lower XXXXX and better XXXXX than lpe', ['computational complexity', 'worst-case performance'])\n", "('method used for task', 'the multi-period XXXXX is extended by considering XXXXX goals', ['vehicle routing problem', 'service consistency'])\n", "('method used for task', 'a 70 percent better XXXXX is achieved by increasing XXXXX by not more than 3.84 percent , on average', ['travel time', 'arrival time consistency'])\n", "('NONE', 'formulated a XXXXX of two XXXXX simultaneously', ['decision problem', 'decision variables'])\n", "('method used for task', 'the impact on distance , XXXXX , and XXXXX are investigated', ['fuel consumption', 'labor costs'])\n", "('method used for task', 'introduction of a multi-attribute XXXXX approach for XXXXX', ['panel data', 'peer assessment'])\n", "('NONE', 'the risk-neutral XXXXX exhibit non-zero XXXXX of variance risk', ['diffusion limits', 'market prices'])\n", "('NONE', 'the risk-neutral XXXXX exhibit non-zero market prices of XXXXX', ['diffusion limits', 'variance risk'])\n", "('NONE', 'the risk-neutral diffusion limits exhibit non-zero XXXXX of XXXXX', ['market prices', 'variance risk'])\n", "('NONE', 'the XXXXX covers 80 XXXXX , 518 threats , and 1244 safeguards', ['knowledge base', 'system components'])\n", "('NONE', 'we present the XXXXX with XXXXX', ['vehicle routing problem', 'release dates'])\n", "('NONE', 'we study XXXXX in an unknown , changing , XXXXX', ['dynamic pricing', 'market environment'])\n", "('NONE', 'XXXXX with quasi-arithmetic means and XXXXX', ['aggregation operators', 'linguistic information'])\n", "('NONE', 'we develop XXXXX ( with XXXXX o ( n2log ( n ) ) for both cases', ['exact algorithms', 'time complexity'])\n", "('method used for task', 'the XXXXX indicator is decomposed into XXXXX ;', ['industry inefficiency', 'sources components'])\n", "('NONE', 'we present a new XXXXX with family setups and XXXXX', ['resource constraints', 'single-machine scheduling problem'])\n", "('method used for task', 'we present a hybrid procedure combining two XXXXX for solving the XXXXX', ['exact methods', 'pricing problem'])\n", "('method used for task', 'our method outperforms an existing branch-and-cut both in XXXXX and XXXXX', ['cpu time', 'solution quality'])\n", "('method used for task', 'group model building is a XXXXX to XXXXX', ['participatory approach', 'system dynamics'])\n", "('method used for task', 'XXXXX is a XXXXX to system dynamics', ['participatory approach', 'group model building'])\n", "('method used for task', 'XXXXX is a participatory approach to XXXXX', ['system dynamics', 'group model building'])\n", "('NONE', 'use the XXXXX with shortages allowed under limited XXXXX', ['eoq model', 'storage capacity'])\n", "('NONE', 'the method used to calculate physical resource tightness is improved by setting a XXXXX of XXXXX', ['critical value', 'resource availability'])\n", "('method used for task', 'the method used to calculate physical XXXXX is improved by setting a XXXXX of resource availability', ['critical value', 'resource tightness'])\n", "('NONE', 'the method used to calculate physical XXXXX is improved by setting a critical value of XXXXX', ['resource availability', 'resource tightness'])\n", "('method used for task', 'an innovative application of XXXXX to analyse hiv XXXXX', ['operations research', 'gene sequences'])\n", "('method used for task', 'XXXXX for XXXXX with 35 , 973 , 840 states were solved', ['optimal control problems', 'production systems'])\n", "('method used for task', 'first bi/multi-objective model for the XXXXX routing and XXXXX', ['home care', 'scheduling problem'])\n", "('method used for task', 'a XXXXX based on multi-directional XXXXX', ['metaheuristic algorithm', 'local search'])\n", "('NONE', 'XXXXX on new XXXXX based on real-life data', ['numerical experiments', 'benchmark instances'])\n", "('method used for task', 'we present the first-ever XXXXX for the consistent XXXXX', ['exact method', 'traveling salesman problem'])\n", "('method used for task', 'we propose novel XXXXX and associated XXXXX', ['valid inequalities', 'separation techniques'])\n", "('NONE', 'XXXXX behaviours are modelled using XXXXX', ['markov models', 'treatment adherence'])\n", "('method used for task', 'we show minimal XXXXX and XXXXX do not lead to minimum costs', ['safety stock', 'inventory variance'])\n", "('method used for task', 'we describe properties of the XXXXX function and compute valid lower and XXXXX', ['upper bounds', 'service time'])\n", "('NONE', 'we examine single-stage dea models with XXXXX that include a very small XXXXX on weights', ['weight restrictions', 'lower bound'])\n", "('NONE', 'we examine single-stage XXXXX with XXXXX that include a very small lower bound on weights', ['weight restrictions', 'dea models'])\n", "('NONE', 'we examine single-stage XXXXX with weight restrictions that include a very small XXXXX on weights', ['lower bound', 'dea models'])\n", "('method used for task', 'green XXXXX programs have many dimensions and XXXXX to consider', ['supplier development', 'big data sets'])\n", "('NONE', 'directions for XXXXX on green XXXXX investment are presented', ['future research', 'supplier development'])\n", "('method used for task', 'minimax regret and XXXXX are used to address XXXXX', ['stochastic programming', 'demand uncertainty'])\n", "('NONE', 'a XXXXX justifies the XXXXX provided by proposed model', ['computational study', 'robust solution'])\n", "('NONE', 'bgeva on woe method for XXXXX showed the best XXXXX', ['missing data', 'predictive accuracy'])\n", "('method used for task', 'a XXXXX for the salesmen is defined by means of a cooperative game theory XXXXX , i.e . the ( least ) core', ['cost allocation', 'solution concept'])\n", "('NONE', 'we model and develop algorithms to reduce XXXXX over XXXXX', ['large-scale networks', 'network size'])\n", "('NONE', 'XXXXX generates more eco-friendly XXXXX under specific conditions', ['supply chain coordination', 'social welfare'])\n", "('method used for task', 'we use XXXXX and XXXXX to solve the problem', ['mixed integer programming', 'benders decomposition'])\n", "('NONE', 'we study the XXXXX of our XXXXX in different setups', ['computation time', 'decomposition algorithm'])\n", "('NONE', 'decision makers’ XXXXX were taken into account in the XXXXX', ['decision process', 'risk preferences'])\n", "('NONE', 'accelerated XXXXX with XXXXX and pareto-optimal cuts', ['benders decomposition', 'valid inequalities'])\n", "('method used for task', 'XXXXX for process mean , pricing , production and XXXXX', ['integrated framework', 'market segmentation'])\n", "('NONE', 'integrating XXXXX with XXXXX in a single framework', ['loss function', 'posterior probability'])\n", "('method used for task', 'using optimized XXXXX and XXXXX', ['monte carlo simulation', 'hybrid genetic algorithm'])\n", "('NONE', 'in a great britain XXXXX , we find that a XXXXX increases generation adequacy', ['case study', 'capacity market'])\n", "('NONE', 'we develop XXXXX of XXXXX to implement ‘stsd’', ['linear systems', 'optimality conditions'])\n", "('method used for task', 'organizations engage in routine XXXXX and XXXXX', ['decision making', 'problem solving'])\n", "('method used for task', 'routine XXXXX and XXXXX have different implications for or', ['decision making', 'problem solving'])\n", "('method used for task', 'our XXXXX based on the branch and price principle solve the multi-trip XXXXX with time windows', ['exact methods', 'vehicle routing problem'])\n", "('method used for task', 'our XXXXX based on the branch and price principle solve the multi-trip vehicle routing problem with XXXXX', ['exact methods', 'time windows'])\n", "('NONE', 'our exact methods based on the branch and price principle solve the multi-trip XXXXX with XXXXX', ['vehicle routing problem', 'time windows'])\n", "('NONE', 'we identify a XXXXX using the XXXXX given by cplex', ['pareto front', 'solution pool'])\n", "('NONE', 'XXXXX with stochastic demands and XXXXX is investigated', ['vehicle routing problem', 'time windows'])\n", "('NONE', 'we show how hard XXXXX can be modeled as XXXXX', ['control problems', 'integer programs'])\n", "('NONE', 'we provide a general framework for searching XXXXX in XXXXX', ['parameter space', 'monte carlo simulation'])\n", "('method used for task', 'XXXXX and lower and XXXXX are added to speed up the algorithm', ['upper bounds', 'dominance rules'])\n", "('NONE', 'the XXXXX outperforms a commercial XXXXX on small instances', ['exact method', 'ilp solver'])\n", "('NONE', 'study dynamic XXXXX with multiple XXXXX in continuous-time', ['portfolio selection', 'risk measures'])\n", "('NONE', 'method is tested using real XXXXX from a london XXXXX', ['road network', 'traffic data'])\n", "('NONE', 'including boosted trees , XXXXX , penalised linear/semi-parametric XXXXX', ['random forests', 'logistic regression'])\n", "('NONE', 'XXXXX suggest boosted XXXXX outperform penalised logistic regression', ['statistical tests', 'regression trees'])\n", "('NONE', 'XXXXX suggest boosted regression trees outperform penalised XXXXX', ['statistical tests', 'logistic regression'])\n", "('NONE', 'statistical tests suggest boosted XXXXX outperform penalised XXXXX', ['regression trees', 'logistic regression'])\n", "('NONE', 'we propose a XXXXX and a comprehensive set of new XXXXX', ['tabu search', 'benchmark instances'])\n", "('NONE', 'an automatic XXXXX as the engine of the dynamic adaptive XXXXX', ['feedback mechanism', 'consensus model'])\n", "('NONE', 'the XXXXX may be replaced by XXXXX', ['standard deviation', 'coherent risk measures'])\n", "('NONE', 'we propose the variable intermediate measures sbm model ( vsbm ) to evaluate the XXXXX of XXXXX', ['system efficiency', 'two-stage processes'])\n", "('method used for task', 'XXXXX to approximate european XXXXX are proposed', ['closed-form formulas', 'option prices'])\n", "('method used for task', 'a XXXXX is applied to XXXXX and u.s. government bond yields', ['calibration procedure', 'option prices'])\n", "('method used for task', 'we use a XXXXX to update the XXXXX under repeated measurements', ['bayesian approach', 'system state'])\n", "('NONE', 'an embedded XXXXX which deals with XXXXX', ['multiple objectives', 'local search procedure'])\n", "('NONE', 'the XXXXX of XXXXX is improved', ['upper bound', 'competitive ratio'])\n", "('method used for task', 'linear-fractional programming and primal-dual XXXXX are used to find the XXXXX of a central inequality', ['upper bound', 'analysis methods'])\n", "('NONE', 'smart and efficient XXXXX composed by auxiliary XXXXX', ['neighborhood structures', 'data structures'])\n", "('NONE', 'ten XXXXX consulting XXXXX', ['system dynamics', 'case studies'])\n", "('method used for task', 'an inverse XXXXX finds a coherent risk measure for a given XXXXX', ['optimal portfolio', 'portfolio problem'])\n", "('method used for task', 'an inverse portfolio problem finds a XXXXX measure for a given XXXXX', ['optimal portfolio', 'coherent risk'])\n", "('NONE', 'an inverse XXXXX finds a XXXXX measure for a given optimal portfolio', ['portfolio problem', 'coherent risk'])\n", "('NONE', 'necessary and XXXXX for the existence of such a XXXXX are obtained', ['sufficient conditions', 'risk measure'])\n", "('method used for task', 'XXXXX ( fs ) is modelled as a ( mixed ) XXXXX', ['feature selection', 'integer optimization problem'])\n", "('NONE', 'the accumulation of biases can cause XXXXX in XXXXX', ['path dependence', 'decision analysis'])\n", "('NONE', 'the framework is XXXXX of XXXXX', ['expected utility', 'profit maximization'])\n", "('NONE', 'autocorrelation , XXXXX , and long order cycles can degrade XXXXX', ['lead time', 'inventory performance'])\n", "('method used for task', 'two new XXXXX are employed to reduce the XXXXX', ['search space', 'dominance rules'])\n", "('method used for task', 'we develop an improved XXXXX based on a XXXXX', ['approximation algorithm', 'minimum spanning tree'])\n", "('method used for task', 'an XXXXX based on a XXXXX is presented for the problem', ['exact algorithm', 'decomposition scheme'])\n", "('method used for task', 'a XXXXX based on a XXXXX is developed', ['heuristic algorithm', 'mixed integer programming'])\n", "('NONE', 'XXXXX disclosed the superiority of XXXXX', ['experimental analysis', 'hybrid algorithms'])\n", "('method used for task', 'cohort XXXXX is applied for solving three XXXXX', ['combinatorial problems', 'intelligence method'])\n", "('NONE', 'we derive a new improved XXXXX of 2.618 for the XXXXX for the online strip packing', ['lower bound', 'competitive ratio'])\n", "('method used for task', 'we derive a new improved XXXXX of 2.618 for the competitive ratio for the online XXXXX', ['lower bound', 'strip packing'])\n", "('method used for task', 'we derive a new improved lower bound of 2.618 for the XXXXX for the online XXXXX', ['competitive ratio', 'strip packing'])\n", "('method used for task', 'we develop a XXXXX based on stochastic dual XXXXX', ['dynamic programming', 'solution approach'])\n", "('method used for task', 'simple XXXXX are summarized by the XXXXX', ['decision rules', 'optimal strategy'])\n", "('method used for task', 'improvements in XXXXX and XXXXX over earlier approaches', ['running time', 'solution quality'])\n", "('NONE', 'decide the rescue team assignment scheme by solving the second XXXXX using the proposed XXXXX', ['heuristic algorithm', 'stage model'])\n", "('NONE', 'network design models augmented with XXXXX provide XXXXX', ['valid inequalities', 'lower bounds'])\n", "('method used for task', 'XXXXX are reported on large XXXXX', ['computational experiments', 'problem instances'])\n", "('NONE', 'the XXXXX involves XXXXX in terms of pairwise comparisons', ['interactive procedure', 'preference information'])\n", "('NONE', 'the XXXXX involves preference information in terms of XXXXX', ['interactive procedure', 'pairwise comparisons'])\n", "('NONE', 'the interactive procedure involves XXXXX in terms of XXXXX', ['preference information', 'pairwise comparisons'])\n", "('NONE', 'generate XXXXX given region-specific characteristics via XXXXX', ['robust solutions', 'decision analysis'])\n", "('method used for task', 'a functional ito’s XXXXX is adopted to evaluate XXXXX', ['calculus approach', 'convex risk measures'])\n", "('method used for task', 'we develop a theory of XXXXX due to XXXXX', ['risk aversion', 'resource dependency'])\n", "('NONE', 'we show that XXXXX significantly impacts optimal XXXXX', ['investment decisions', 'resource dependency'])\n", "('NONE', 'uncertainty and XXXXX are XXXXX for resource dependency', ['risk aversion', 'sufficient conditions'])\n", "('method used for task', 'uncertainty and XXXXX are sufficient conditions for XXXXX', ['risk aversion', 'resource dependency'])\n", "('method used for task', 'uncertainty and risk aversion are XXXXX for XXXXX', ['sufficient conditions', 'resource dependency'])\n", "('NONE', 'we formulate and solve the pickup and XXXXX with XXXXX and multiple stacks', ['time windows', 'delivery problem'])\n", "('NONE', 'two branch-price-and-cut algorithms are implemented to solve the pickup and XXXXX with XXXXX and multiple stacks', ['time windows', 'delivery problem'])\n", "('method used for task', 'new XXXXX and XXXXX are developed', ['lower bounds', 'dominance rules'])\n", "('method used for task', 'a XXXXX is suggested to transform XXXXX into a linear one', ['heuristic algorithm', 'nonlinear model'])\n", "('NONE', 'we model a XXXXX with XXXXX from competitors', ['competitive environment', 'incomplete data'])\n", "('method used for task', 'the assignment of decision objects to decision classes is based on the use of the XXXXX and XXXXX', ['majority principle', 'veto effect'])\n", "('method used for task', 'the XXXXX and XXXXX notions are implemented through the concepts of concordance and discordance', ['majority principle', 'veto effect'])\n", "('NONE', 'coelli et al.’s ( 2007 ) environmental XXXXX does not reward XXXXX', ['efficiency measure', 'pollution control'])\n", "('NONE', 'a new environmental XXXXX that rewards XXXXX efforts is proposed', ['efficiency measure', 'pollution control'])\n", "('NONE', 'a XXXXX identifies cases solvable in XXXXX', ['complexity analysis', 'polynomial time'])\n", "('NONE', 'item rearrangements along XXXXX bring significant routing XXXXX', ['vehicle routes', 'cost savings'])\n", "('method used for task', 'we examine a parallel machine scheduling problem with XXXXX and XXXXX', ['setup times', 'time windows'])\n", "('method used for task', 'XXXXX ( ip ) and XXXXX ( cp ) models are proposed', ['integer programming', 'constraint programming'])\n", "('NONE', 'the XXXXX of the XXXXX is on https : //github.com/ctu-iig/nrrpgpu', ['source code', 'parallel algorithm'])\n", "('NONE', 'we present a XXXXX based XXXXX', ['benders decomposition', 'solution method'])\n", "('NONE', 'XXXXX when dual XXXXX outperforms single bundle price', ['numerical analysis', 'pricing strategy'])\n", "('NONE', 'using a XXXXX , our model can be solved with a standard XXXXX', ['decomposition approach', 'ip solver'])\n", "('method used for task', 'this XXXXX shock models based on a new XXXXX of shocks', ['counting process', 'paper studies'])\n", "('method used for task', 'XXXXX and consistency of the mle is established for XXXXX', ['asymptotic normality', 'mle metamodels'])\n", "('NONE', 'we study XXXXX and advertising policies in a XXXXX for a monopolist firm', ['dynamic pricing', 'sales model'])\n", "('method used for task', 'we consider the XXXXX and XXXXX in two machine flow shops', ['order acceptance', 'scheduling problem'])\n", "('method used for task', 'we use XXXXX to model spatial/functional connectivity in XXXXX', ['integer programming', 'site selection'])\n", "('method used for task', 'we present a XXXXX for the size-robust XXXXX', ['knapsack problem', 'column generation algorithm'])\n", "('method used for task', 'we present a XXXXX for the demand robust XXXXX', ['shortest path problem', 'column generation algorithm'])\n", "('method used for task', 'we show that XXXXX decomposes uniquely into technical efficiency change and XXXXX', ['technical change', 'productivity change'])\n", "('NONE', 'we derive an aggregate XXXXX from individual XXXXX', ['productivity index', 'productivity indices'])\n", "('NONE', 'XXXXX based XXXXX on the quality of service are devised', ['mathematical programming', 'lower bounds'])\n", "('NONE', 'we develop novel heuristics for XXXXX in XXXXX', ['channel assignment', 'wireless mesh networks'])\n", "('method used for task', 'we review the recent advances in XXXXX for XXXXX , minlp', ['global optimization', 'mixed integer nonlinear programming'])\n", "('method used for task', 'an XXXXX is applied to a XXXXX with profits', ['adaptive large neighborhood search', 'vehicle routing problem'])\n", "('method used for task', 'the XXXXX were successfully applied to a XXXXX and their worth is demonstrated', ['case study', 'expansion models'])\n", "('NONE', 'XXXXX : all phases of XXXXX', ['disaster management', 'application area'])\n", "('NONE', 'XXXXX are classified and discussed , as well as XXXXX', ['objective functions', 'solution methods'])\n", "('NONE', 'we consider a XXXXX with XXXXX on unrelated parallel machines', ['scheduling model', 'maintenance activities'])\n", "('NONE', 'analytic and XXXXX of the two-stage XXXXX are presented', ['numerical solutions', 'dynamic model'])\n", "('method used for task', 'q-g and q-cg are q-analogs of the XXXXX and XXXXX', ['steepest descent', 'conjugate gradient methods'])\n", "('NONE', 'an efficient XXXXX that considers the XXXXX and “solvency-based” measures to produce offspring', ['scatter search', 'search history'])\n", "('NONE', 'we use a XXXXX to estimate the distribution of XXXXX times between any two locations in a city', ['bayesian method', 'ambulance travel'])\n", "('method used for task', 'our XXXXX estimates outperform a recently-published method and a commercial XXXXX', ['travel time', 'software package'])\n", "('NONE', 'the XXXXX of nonlinear XXXXX is continuous with respect to changing the measure', ['value function', 'stochastic optimization problems'])\n", "('method used for task', 'algorithm to minimize XXXXX for tree-shaped XXXXX', ['investment cost', 'gas distribution networks'])\n", "('method used for task', 'we extend the XXXXX to take into account a XXXXX of the set of criteria', ['hierarchical structure', 'electre tri methods'])\n", "('method used for task', 'the XXXXX hierarchy process is applied in conjunction with the XXXXX', ['multiple criteria', 'electre tri methods'])\n", "('method used for task', 'two XXXXX based on johnsons rule and XXXXX', ['approximation algorithms', 'linear programming'])\n", "('NONE', 'XXXXX of 4 XXXXX of pmp is conducted', ['comparative study', 'exact solution methods'])\n", "('method used for task', 'develop algorithm to compute XXXXX for discrete XXXXX', ['robust solutions', 'optimisation problems'])\n", "('NONE', 'we present a cyclic production scheme for XXXXX with XXXXX', ['multiple products', 'stochastic demand'])\n", "('NONE', 'we consider XXXXX , XXXXX and storage capacity', ['sequence-dependent setup times', 'service levels'])\n", "('method used for task', 'we consider XXXXX , service levels and XXXXX', ['sequence-dependent setup times', 'storage capacity'])\n", "('method used for task', 'we consider sequence-dependent setup times , XXXXX and XXXXX', ['service levels', 'storage capacity'])\n", "('NONE', 'the XXXXX based on real world data compares six strategies that control the XXXXX', ['computational study', 'cycle length'])\n", "('NONE', 'we present a biobjective formulation for identifying XXXXX from a given XXXXX', ['robust solutions', 'pareto set'])\n", "('method used for task', 'XXXXX and a solution algorithm are developed for the case of multiobjective XXXXX', ['structural properties', 'linear programs'])\n", "('method used for task', 'XXXXX and a XXXXX are developed for the case of multiobjective linear programs', ['structural properties', 'solution algorithm'])\n", "('NONE', 'structural properties and a XXXXX are developed for the case of multiobjective XXXXX', ['linear programs', 'solution algorithm'])\n", "('method used for task', 'we derive equivalent XXXXX . thus , we compute optimal solutions and XXXXX', ['upper bounds', 'mixed integer linear programming formulations'])\n", "('NONE', 'a XXXXX illustrates XXXXX of defensive resources', ['case study', 'optimal allocation'])\n", "('method used for task', 'we take a XXXXX and develop a XXXXX', ['branch-and-price algorithm', 'column generation approach'])\n", "('NONE', 'we provide a competitive XXXXX of the online one-way trading problem with XXXXX on prices', ['limited information', 'difference analysis'])\n", "('NONE', 'the gas market is represented by a partial XXXXX including XXXXX', ['market power', 'equilibrium model'])\n", "('NONE', 'we develop a XXXXX embedding fairness concerns into the XXXXX', ['behavioral model', 'utility functions'])\n", "('method used for task', 'we propose a novel XXXXX , considering unloading activities at the terminal and XXXXX scheduling', ['integrated model', 'pipeline pumping'])\n", "('NONE', 'we propose a methodology based on XXXXX , XXXXX and a method to obtain optimal scenario trees', ['historical data', 'statistical analysis'])\n", "('method used for task', 'we propose a methodology based on XXXXX , statistical analysis and a method to obtain optimal XXXXX', ['historical data', 'scenario trees'])\n", "('method used for task', 'we propose a methodology based on historical data , XXXXX and a method to obtain optimal XXXXX', ['statistical analysis', 'scenario trees'])\n", "('method used for task', 'XXXXX for XXXXX at isolated intersections are derived', ['stochastic models', 'queue lengths'])\n", "('method used for task', 'we analyze XXXXX for XXXXX projects , accounting for market risk preferences', ['investment timing', 'risk reduction'])\n", "('NONE', 'a mechanism to discard low XXXXX before the XXXXX is proposed', ['local search', 'quality solutions'])\n", "('NONE', 'combination of binary XXXXX and minimization of XXXXX', ['decision trees', 'boolean functions'])\n", "('method used for task', 'identifies the selection of XXXXX for XXXXX as research question', ['business analytics', 'data items'])\n", "('NONE', 'we model the XXXXX problem with XXXXX for container departure', ['time windows', 'container relocation'])\n", "('NONE', 'our approach accounts for dependencies among XXXXX and uncertainty in XXXXX', ['project scores', 'portfolio constraints'])\n", "('method used for task', 'equivalence of surcharge pricing with joint economic XXXXX and XXXXX has been established', ['lot sizing', 'quantity discount'])\n", "('method used for task', 'we study the two-machine flowshop problem with sequence-independent XXXXX to minimize XXXXX', ['setup times', 'total completion time'])\n", "('NONE', 'we study the XXXXX with sequence-independent XXXXX to minimize total completion time', ['setup times', 'two-machine flowshop problem'])\n", "('method used for task', 'we study the XXXXX with sequence-independent setup times to minimize XXXXX', ['total completion time', 'two-machine flowshop problem'])\n", "('NONE', 'XXXXX assess the efficiency of XXXXX', ['numerical experiments', 'branch-and-bound algorithms'])\n", "('method used for task', 'we present a XXXXX for XXXXX', ['supplier selection', 'stochastic programming model'])\n", "('NONE', 'solving a single XXXXX in the XXXXX is more efficient', ['benders decomposition', 'master problem'])\n", "('NONE', 'we examine the effects of service and XXXXX on XXXXX', ['optimal policy', 'competition intensity'])\n", "('method used for task', 'we propose an XXXXX for an XXXXX', ['exact algorithm', 'np-hard problem'])\n", "('method used for task', 'we characterize the structure of its XXXXX and derive XXXXX', ['optimal policy', 'analytical solution'])\n", "('method used for task', 'a XXXXX for XXXXX and economic manufacturing quantity is proposed', ['condition-based maintenance', 'joint optimization model'])\n", "('NONE', 'the assumption that the XXXXX has discrete XXXXX has been relaxed', ['state space', 'degradation process'])\n", "('NONE', 'the XXXXX from the XXXXX are considerable when the costs related to unexpected failures are high', ['cost savings', 'joint optimization model'])\n", "('method used for task', 'we study XXXXX where the processing times are also XXXXX', ['scheduling problems', 'decision variables'])\n", "('method used for task', 'co-opetition leads to more profit and less XXXXX based on higher XXXXX and increased unit XXXXX', ['carbon emissions', 'product prices'])\n", "('method used for task', 'XXXXX is used to find the minimum XXXXX', ['multiobjective optimization', 'separation values'])\n", "('NONE', 'presented a risk shaping model for XXXXX under XXXXX', ['random yield', 'production planning problem'])\n", "('NONE', 'used XXXXX as the XXXXX', ['conditional value-at-risk', 'risk measure'])\n", "('NONE', 'XXXXX increases price , whereas XXXXX reduces it', ['demand uncertainty', 'cost uncertainty'])\n", "('NONE', 'this paper raises the problem of measuring XXXXX of XXXXX in dea models', ['partial derivatives', 'transformation functions'])\n", "('NONE', 'this paper raises the problem of measuring XXXXX of transformation functions in XXXXX', ['partial derivatives', 'dea models'])\n", "('NONE', 'this paper raises the problem of measuring partial derivatives of XXXXX in XXXXX', ['transformation functions', 'dea models'])\n", "('method used for task', 'we propose a new XXXXX for solving multi-objective XXXXX', ['exact algorithm', 'integer programs'])\n", "('NONE', 'analysis the impact of XXXXX on XXXXX', ['capacity planning', 'problem characteristics'])\n", "('method used for task', 'we present a XXXXX ( de ) for XXXXX', ['modified differential evolution', 'dynamic optimization'])\n", "('NONE', 'results show that the XXXXX can exhibit chaos when XXXXX are at work', ['social interactions', 'demand dynamics'])\n", "('method used for task', 'in numerical study , we show that the XXXXX is superior to the traditional XXXXX', ['empirical bayes method', 'estimation method'])\n", "('method used for task', 'a heuristic XXXXX for this new XXXXX was developed', ['optimization problem', 'solution method'])\n", "('method used for task', 'we analyze XXXXX for the erlang-a XXXXX', ['risk measures', 'queueing model'])\n", "('NONE', 'we analyze XXXXX of queues using XXXXX', ['performance measures', 'risk measures'])\n", "('method used for task', 'we examine the effect of XXXXX and XXXXX on four types of supply contracts', ['market share', 'asymmetric information'])\n", "('NONE', 'the optimal transfer payment of a buyer is influenced by XXXXX of the buyer and XXXXX', ['market share', 'supply network structure'])\n", "('method used for task', 'uncertainty in XXXXX and the underlying XXXXX is modeled', ['probability distribution', 'risk preferences'])\n", "('method used for task', 'the XXXXX is computed exactly with XXXXX in a special case', ['optimal solution', 'convex optimization'])\n", "('method used for task', 'XXXXX are conducted for the newsvendor and XXXXX to analyze the implications of our model', ['numerical experiments', 'portfolio optimization problem'])\n", "('NONE', 'we investigate the XXXXX of XXXXX and forward-start options', ['model risk', 'variance swaps'])\n", "('NONE', 'we provide evidence that the XXXXX of XXXXX is around 100 basis points', ['model risk', 'variance swaps'])\n", "('method used for task', 'we propose a XXXXX to determine open-loop XXXXX', ['nash equilibria', 'search scheme'])\n", "('NONE', 'we give a new definition of the XXXXX inspired by the weak link in XXXXX', ['system efficiency', 'supply chains'])\n", "('NONE', 'the XXXXX a three-player XXXXX with an infinite time horizon', ['differential game', 'paper studies'])\n", "('NONE', 'the paper studies a three-player XXXXX with an infinite XXXXX', ['differential game', 'time horizon'])\n", "('NONE', 'the XXXXX a three-player differential game with an infinite XXXXX', ['paper studies', 'time horizon'])\n", "('method used for task', 'we propose a new XXXXX for solving tri-objective XXXXX', ['exact algorithm', 'integer programs'])\n", "('NONE', 'several unique XXXXX of an XXXXX are proved', ['structural properties', 'optimal schedule'])\n", "('method used for task', 'a fork-join XXXXX is formulated to analyze the XXXXX', ['queueing network', 'system performance'])\n", "('NONE', 'the XXXXX has advantages in small XXXXX', ['parallel policy', 'size systems'])\n", "('method used for task', 'we propose an XXXXX with efficient heuristics to explore the XXXXX', ['adaptive large neighborhood search', 'solution space'])\n", "('NONE', 'we provide an improved wolf XXXXX ( wsa ) using a global XXXXX', ['search algorithm', 'memory structure'])\n", "('method used for task', 'XXXXX for XXXXX and generation capacity investment', ['transmission line', 'equilibrium model'])\n", "('NONE', 'we perform a XXXXX on instances adapted from vrp XXXXX', ['computational study', 'benchmark instances'])\n", "('method used for task', 'the effect of XXXXX on the maximum XXXXX is analyzed', ['stochastic demand', 'inventory distribution'])\n", "('method used for task', 'the XXXXX take past , present and future XXXXX into account', ['performance indicators', 'dea models'])\n", "('method used for task', 'we propose a XXXXX for this novel XXXXX', ['objective function', 'dynamic pricing model'])\n", "('NONE', 'we study channel XXXXX of a dual-channel XXXXX facing potential supply shortage', ['supply chain', 'priority strategy'])\n", "('NONE', 'we examine the effects of coordination/decision sequence of XXXXX on XXXXX', ['channel priority', 'priority strategy'])\n", "('method used for task', 'pricing and XXXXX strategies are robust to the time sequence of XXXXX decision', ['channel priority', 'channel strategies'])\n", "('method used for task', 'pricing and XXXXX strategies are robust to the time sequence of XXXXX decision', ['channel priority', 'channel decision'])\n", "('method used for task', 'pricing and XXXXX strategies are robust to the time sequence of XXXXX decision', ['channel strategies', 'channel priority'])\n", "('method used for task', 'pricing and XXXXX strategies are robust to the time sequence of XXXXX decision', ['channel priority', 'channel decision'])\n", "('method used for task', 'this XXXXX the optimal preventive maintenance problem based on a new XXXXX', ['counting process', 'paper studies'])\n", "('method used for task', 'we study admission and XXXXX for a make-to-order XXXXX', ['inventory control', 'production system'])\n", "('NONE', 'we investigate the XXXXX under non-preemptive XXXXX and setup cost', ['optimal policy', 'lead time'])\n", "('method used for task', 'we investigate the XXXXX under non-preemptive lead time and XXXXX', ['optimal policy', 'setup cost'])\n", "('method used for task', 'we investigate the optimal policy under non-preemptive XXXXX and XXXXX', ['lead time', 'setup cost'])\n", "('method used for task', 'a modified wordnet based XXXXX for XXXXX disambiguation', ['similarity measure', 'word sense'])\n", "('method used for task', 'two new XXXXX are proposed based on joint XXXXX', ['mutual information', 'feature selection methods'])\n", "('method used for task', 'the best XXXXX and the lowest XXXXX are achieved', ['time complexity', 'prediction performance'])\n", "('NONE', 'uncertain XXXXX can be included in the XXXXX for aerospace panels', ['boundary conditions', 'design process'])\n", "('method used for task', 'we propose a level set-based XXXXX to design acoustic metamaterials using the XXXXX', ['topology optimization', 'finite element method'])\n", "('NONE', 'we formulate the XXXXX using effective XXXXX', ['optimization problem', 'material properties'])\n", "('NONE', 'a two-step XXXXX enabled to handle the design of effective XXXXX', ['optimization process', 'bulk modulus'])\n", "('method used for task', 'algorithms for calculating the XXXXX and the internal XXXXX are presented', ['stiffness matrix', 'force vector'])\n", "('method used for task', 'we address an isogeometric XXXXX for thermal XXXXX of functionally graded material plates', ['finite element formulation', 'buckling analysis'])\n", "('NONE', 'analysis of plate and XXXXX may involve inclined principal XXXXX', ['shell structures', 'stress fields'])\n", "('method used for task', 'a new XXXXX for steel–concrete side-plated beams in XXXXX is presented', ['finite element model', 'fire conditions'])\n", "('method used for task', 'the proposed element has a simple formulation meanwhile it shows satisfactory XXXXX and possesses XXXXX', ['high accuracy', 'convergence performance'])\n", "('method used for task', 'a refined 18-dof triangular hybrid XXXXX is given out , and the element captures the scale effect and has XXXXX', ['high accuracy', 'stress element'])\n", "('method used for task', 'interpolet-based XXXXX are obtained and presented graphically . expressions for XXXXX and force vector are derived', ['shape functions', 'stiffness matrix'])\n", "('method used for task', 'interpolet-based XXXXX are obtained and presented graphically . expressions for stiffness matrix and XXXXX are derived', ['shape functions', 'force vector'])\n", "('method used for task', 'interpolet-based shape functions are obtained and presented graphically . expressions for XXXXX and XXXXX are derived', ['stiffness matrix', 'force vector'])\n", "('NONE', 'XXXXX are represented as discrete point load at the XXXXX', ['muscle forces', 'integration points'])\n", "('NONE', 'XXXXX enables the use of XXXXX , fully integrated solid tetrahedral finite elements', ['parallel processing', 'high order'])\n", "('NONE', 'XXXXX enables the use of high order , fully integrated solid tetrahedral XXXXX', ['parallel processing', 'finite elements'])\n", "('NONE', 'parallel processing enables the use of XXXXX , fully integrated solid tetrahedral XXXXX', ['high order', 'finite elements'])\n", "('NONE', 'evolutionary XXXXX are derived from XXXXX', ['experimental measurements', 'power spectra'])\n", "('NONE', 'condensation technique of degree of freedom is first proposed to improve the XXXXX of XXXXX with galerkin weak form', ['computational efficiency', 'meshfree method'])\n", "('NONE', 'adoption of modified crack closure integral ( mcci ) in XXXXX for evaluation of stress XXXXX ( sifs )', ['meshfree methods', 'intensity factors'])\n", "('method used for task', 'XXXXX based on XXXXX is consistent with the field results', ['fe modal analysis', 'vehicle waveforms'])\n", "('method used for task', 'XXXXX based on XXXXX can be used to guide bridge inspectors', ['fe modal analysis', 'vehicle waveforms'])\n", "('NONE', 'XXXXX is simplified through a number of observed XXXXX', ['model construction', 'statistical properties'])\n", "('NONE', 'calculate the exact XXXXX between different stress modes using our new XXXXX', ['inner product', 'similarity degrees'])\n", "('NONE', 'calculate the exact similarity degrees between different XXXXX using our new XXXXX', ['inner product', 'stress modes'])\n", "('NONE', 'calculate the exact XXXXX between different XXXXX using our new inner product', ['similarity degrees', 'stress modes'])\n", "('NONE', 'derive the basic XXXXX from XXXXX and broken them into a set of sub-modes', ['displacement field', 'stress modes'])\n", "('method used for task', 'XXXXX is a method called numerical microscope in XXXXX and numerical analysis', ['wavelet analysis', 'signal processing'])\n", "('method used for task', 'XXXXX is a method called numerical microscope in signal processing and XXXXX', ['wavelet analysis', 'numerical analysis'])\n", "('method used for task', 'wavelet analysis is a method called numerical microscope in XXXXX and XXXXX', ['signal processing', 'numerical analysis'])\n", "('NONE', 'an efficient XXXXX of 3d semi-rigid XXXXX is proposed', ['steel frames', 'analysis procedure'])\n", "('method used for task', 'a XXXXX to address uncertainty effects in the XXXXX is presented', ['hybrid method', 'impact force identification'])\n", "('method used for task', 'a XXXXX is conducted to analyze the effects of material uncertainty on XXXXX of composite stiffened panels', ['parametric study', 'dynamic responses'])\n", "('NONE', 'influence of the XXXXX kinetics was taken into account by implementing in the fem model XXXXX data', ['phase transformation', 'phase transformation kinetics'])\n", "('NONE', 'XXXXX procedure at XXXXX is based on diffuse approximation', ['field transfer', 'integration points'])\n", "('NONE', 'it has a correct rank , is free from XXXXX , with no signs of XXXXX', ['spurious modes', 'shear locking'])\n", "('NONE', 'thermal behavior coupled with XXXXX in comprehensive XXXXX', ['structural analysis', 'fe model'])\n", "('method used for task', 'similar to the end shortening strain commonly used for compression , the skewed XXXXX is uniquely proposed for the first time to control the in-plane shear action in the finite XXXXX', ['angle strain', 'strip method'])\n", "('method used for task', 'when the ratio of skewed XXXXX vs. end shortening strain is large enough , the average tensile XXXXX along the longitudinal direction are produced due to the large out-of-plane deflection', ['angle strain', 'section forces'])\n", "('method used for task', 'a new XXXXX for XXXXX shear deformable composite beams is proposed', ['finite element', 'higher order'])\n", "('method used for task', 'a new XXXXX for higher order shear deformable XXXXX is proposed', ['finite element', 'composite beams'])\n", "('NONE', 'a new finite element for XXXXX shear deformable XXXXX is proposed', ['higher order', 'composite beams'])\n", "('method used for task', 'XXXXX and XXXXX can be used to calculate fracture toughness', ['exact method', 'finite element'])\n", "('method used for task', 'XXXXX and finite element can be used to calculate XXXXX', ['exact method', 'fracture toughness'])\n", "('method used for task', 'exact method and XXXXX can be used to calculate XXXXX', ['finite element', 'fracture toughness'])\n", "('method used for task', 'the framework is based on coupled XXXXX and XXXXX', ['genetic algorithm', 'finite element analysis'])\n", "('method used for task', 'XXXXX and XXXXX , insensitive mesh distortions , free volume locking , etc . are found for the cq4', ['high accuracy', 'convergence rate'])\n", "('NONE', 'XXXXX show the good performance of the proposed XXXXX', ['numerical examples', 'multiscale method'])\n", "('method used for task', 'XXXXX based on the laplace transform method are provided for the XXXXX', ['analytical solutions', 'case studies'])\n", "('method used for task', 'the XXXXX is used to construct a material XXXXX', ['density approximation', 'shepard function'])\n", "('NONE', 'XXXXX formulated as independent from angle and number of XXXXX', ['stiffness matrix', 'fourier series'])\n", "('method used for task', 'a new hybrid inactive/quiet XXXXX is proposed for modeling XXXXX', ['additive manufacturing', 'element method'])\n", "('NONE', 'XXXXX of hydrostatic XXXXX', ['performance characteristics', 'thrust bearing'])\n", "('NONE', 'XXXXX of parallel XXXXX independent of element order', ['computational time', 'matrix factorization'])\n", "('method used for task', 'fundamental XXXXX cases are investigated for XXXXX welded aluminum joints', ['impact analysis', 'friction stir'])\n", "('NONE', 'we investigate a deterministic and statistical XXXXX in XXXXX', ['size effect', 'reinforced concrete beams'])\n", "('method used for task', 'we propose a XXXXX based on the XXXXX', ['finite element formulation', 'discontinuous galerkin method'])\n", "('NONE', 'a consistent XXXXX of the nonlocal XXXXX is presented', ['variational formulation', 'timoshenko beam'])\n", "('method used for task', 'use of volumetric XXXXX and pressure enhancement strategies for low XXXXX', ['strain rate', 'order finite elements'])\n", "('method used for task', 'six XXXXX are provided to investigate the strengths of the method through the XXXXX', ['numerical examples', 'software application'])\n", "('NONE', 'regular grid insertion to achieve XXXXX in XXXXX', ['linear complexity', 'delaunay triangulation'])\n", "('method used for task', 'use of a XXXXX and a XXXXX at two different stages of the same computation', ['3d model', 'beam model'])\n", "('NONE', 'a spline XXXXX constructed in practical domain can be dealt with as a XXXXX', ['meshless method', 'element method'])\n", "('NONE', 'a XXXXX with commercial XXXXX is developed to figure out the constraint equations', ['numerical method', 'fe codes'])\n", "('NONE', 'this fe has no numerical pathology ( XXXXX , XXXXX , … )', ['shear locking', 'spurious modes'])\n", "('NONE', 'XXXXX of higher-order XXXXX at interfaces', ['numerical implementation', 'boundary conditions'])\n", "('NONE', 'XXXXX gradients are calculated using weighted XXXXX', ['meshless method', 'plastic strain'])\n", "('NONE', 'a selected number of quasi-static and dynamic XXXXX has been chosen to show the performance of the new XXXXX', ['numerical examples', 'contact formulation'])\n", "('method used for task', 'several XXXXX operators between meshes are studied for XXXXX and remeshing', ['ale formulation', 'field transfer'])\n", "('NONE', 'a formulation of a 2d XXXXX with parabolic through-the-thickness XXXXX was carried out', ['temperature distribution', 'heat transfer element'])\n", "('method used for task', 'three XXXXX were developed to simulate XXXXX in pre-tensioned concrete and validated against previous experiments', ['numerical models', 'prestress transfer'])\n", "('method used for task', 'a XXXXX was conducted on factors affecting the XXXXX', ['parametric study', 'prestress transfer'])\n", "('method used for task', 'a meshfree interface-finite XXXXX ( mi-fem ) for XXXXX', ['phase transformation', 'element method'])\n", "('method used for task', 'solving XXXXX for strain gradient XXXXX', ['mixed finite element formulation', 'elasticity problems'])\n", "('method used for task', 'a novel and simple fsdt-based XXXXX for XXXXX of fgm plates is presented', ['isogeometric analysis', 'geometrically nonlinear analysis'])\n", "('NONE', 'simplified XXXXX element formulation satisfying a priori the XXXXX', ['finite strain', 'patch test'])\n", "('method used for task', 'the proposed XXXXX is based on the XXXXX', ['optimization procedure', 'beso method'])\n", "('NONE', 'the objective is to minimize the XXXXX of the XXXXX', ['frequency response', 'coupled systems'])\n", "('NONE', 'XXXXX of topologically complex XXXXX is performed by using trimmed nurbs surfaces and exact normal vectors', ['isogeometric analysis', 'shell structures'])\n", "('NONE', 'XXXXX of topologically complex shell structures is performed by using trimmed XXXXX and exact normal vectors', ['isogeometric analysis', 'nurbs surfaces'])\n", "('method used for task', 'isogeometric analysis of topologically complex XXXXX is performed by using trimmed XXXXX and exact normal vectors', ['shell structures', 'nurbs surfaces'])\n", "('method used for task', 'the exact normal vectors , which are directly calculated by XXXXX are used in the shell formulation based on the reissner–mindlin degenerated XXXXX', ['shell element', 'nurbs surface expression'])\n", "('method used for task', 'the exact normal vectors , which are directly calculated by nurbs surface expression are used in the XXXXX based on the reissner–mindlin degenerated XXXXX', ['shell element', 'shell formulation'])\n", "('NONE', 'the exact normal vectors , which are directly calculated by XXXXX are used in the XXXXX based on the reissner–mindlin degenerated shell element', ['nurbs surface expression', 'shell formulation'])\n", "('NONE', 'the analytic derivatives of the direction vectors , which are directly calculated by XXXXX are also employed in the XXXXX', ['nurbs surface expression', 'shell formulation'])\n", "('method used for task', 'with XXXXX , it is shown that the present method for topologically complex XXXXX , which can not be treated by the conventional isogeometric analysis without multiple patches , gives appropriate solutions', ['numerical examples', 'shell structures'])\n", "('method used for task', 'with XXXXX , it is shown that the present method for topologically complex shell structures , which can not be treated by the conventional XXXXX without multiple patches , gives appropriate solutions', ['numerical examples', 'isogeometric analysis'])\n", "('NONE', 'with numerical examples , it is shown that the present method for topologically complex XXXXX , which can not be treated by the conventional XXXXX without multiple patches , gives appropriate solutions', ['shell structures', 'isogeometric analysis'])\n", "('NONE', 'a simple XXXXX for prediction of buckled XXXXX is proposed', ['analytical model', 'beam stiffness'])\n", "('NONE', 'XXXXX of the XXXXX is achieved using a model reduction technique', ['fast computation', 'wave modes'])\n", "('NONE', 'large XXXXX are obtained when computing the forced response of XXXXX', ['periodic structures', 'cpu time savings'])\n", "('NONE', 'established the effectiveness of the three-phase XXXXX approach in capturing the non-linearity of the materials considered without sacrificing the simplicity of the XXXXX model', ['unit cell', 'unit cell approach'])\n", "('NONE', 'different XXXXX were also conducted to identify the influence of certain phase characteristics of the XXXXX for the materials considered', ['parametric studies', 'unit cell'])\n", "('NONE', 'XXXXX of the multiscale XXXXX are derived on some regularity hypothesis', ['error estimates', 'approximate solution'])\n", "('NONE', 'XXXXX enables an XXXXX of fatigue stress–strain loop', ['efficient computation', 'control protocol'])\n", "('method used for task', 'XXXXX are successfully tested for XXXXX and structures', ['control algorithms', 'volume elements'])\n", "('method used for task', 'meshing procedures have a significant effect on the XXXXX and the von mises stress ( XXXXX vs. brick mesh )', ['tetrahedral mesh', 'fragmentation behavior'])\n", "('NONE', 'second-order effects are considered by XXXXX , save XXXXX', ['computational time', 'stability functions'])\n", "('NONE', 'first static and dynamic study of lsfes for rm XXXXX on XXXXX', ['composite plates', 'unstructured grids'])\n", "('NONE', 'first comparison of rm lsfes with shell281 elements in terms of XXXXX , XXXXX', ['cpu time', 'model size'])\n", "('method used for task', 'the proposed XXXXX is based on the XXXXX', ['optimization procedure', 'beso method'])\n", "('method used for task', 'the virtual-surface XXXXX , which is based on inside–outside method and penalty function method , is developed . innermost details regarding XXXXX , contact force calculation and the algorithmic implementations are presented', ['local search', 'contact algorithm'])\n", "('method used for task', 'in the proposed contact algorithm , the penetration between XXXXX and XXXXX can be determined without solving nonlinear equations', ['discrete element', 'finite element'])\n", "('NONE', 'in the proposed XXXXX , the penetration between XXXXX and finite element can be determined without solving nonlinear equations', ['discrete element', 'contact algorithm'])\n", "('method used for task', 'in the proposed XXXXX , the penetration between discrete element and XXXXX can be determined without solving nonlinear equations', ['finite element', 'contact algorithm'])\n", "('NONE', 'the proposed XXXXX has been implemented into in-house developed code . four XXXXX are employed to validate the accuracy and robustness of the proposed XXXXX', ['numerical examples', 'contact algorithm'])\n", "('NONE', 'the proposed XXXXX has been implemented into in-house developed code . four XXXXX are employed to validate the accuracy and robustness of the proposed XXXXX', ['numerical examples', 'contact algorithm'])\n", "('NONE', 'buckling and XXXXX of dimpled XXXXX were investigated', ['ultimate strength', 'steel columns'])\n", "('method used for task', 'extension of previous work of the authors on constitutive contact laws in combination with the dual mortar method for XXXXX to quadratic XXXXX', ['contact problems', 'finite elements'])\n", "('method used for task', 'suggestion of a petrov galerkin dual mortar approach for quadratic XXXXX to avoid XXXXX in case of large curvatures or high gradients', ['finite elements', 'consistency errors'])\n", "('method used for task', 'a mode patch that makes the algorithm applicable to both XXXXX and three-dimensional XXXXX', ['plane stress', 'stress states'])\n", "('NONE', 'the XXXXX permits the use of gradient-based methods in XXXXX in axisymmetry with torsional loads', ['sensitivity analysis', 'shape optimization'])\n", "('method used for task', 'an efficient XXXXX is developed for XXXXX of 3d bi-modulus materials', ['computational method', 'geometrically nonlinear analysis'])\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "('method used for task', 'the proposed model provide high-precision XXXXX and local XXXXX', ['load distribution', 'stress field'])\n", "('method used for task', 'the new fe model considers both XXXXX and size effects for micro XXXXX of circular cups', ['surface roughness', 'deep drawing'])\n", "('NONE', 'the new XXXXX considers both XXXXX and size effects for micro deep drawing of circular cups', ['surface roughness', 'fe model'])\n", "('method used for task', 'the new XXXXX considers both surface roughness and size effects for micro XXXXX of circular cups', ['deep drawing', 'fe model'])\n", "('NONE', 'XXXXX affects the springback , the drawability and the cups’ quality obviously in micro XXXXX', ['surface roughness', 'deep drawing'])\n", "('method used for task', 'the XXXXX is given by a residual-stress dependent nonlinear elastic XXXXX in terms of invariants', ['material model', 'constitutive law'])\n", "('method used for task', 'the dependence of bifurcation and postbifurcation behavior of tubes under torsion on XXXXX is illustrated and compared with results when there is no XXXXX', ['residual stresses', 'residual stress'])\n", "('NONE', 'XXXXX of the coupled governing XXXXX', ['differential equations', 'solution strategy'])\n", "('NONE', 'a method is proposed to blend nurbs and XXXXX in 3d XXXXX', ['finite elements', 'isogeometric analysis'])\n", "('method used for task', 'a mesh-adaptive XXXXX has been implemented to work along with the non-local XXXXX', ['finite element technique', 'damage model'])\n", "('method used for task', 'development of a novel XXXXX for the combined finite-discrete XXXXX by means of ghost particles', ['coupling approach', 'element method'])\n", "('NONE', 'development of a novel XXXXX for the combined finite-discrete element method by means of XXXXX', ['coupling approach', 'ghost particles'])\n", "('NONE', 'development of a novel coupling approach for the combined finite-discrete XXXXX by means of XXXXX', ['element method', 'ghost particles'])\n", "('method used for task', 'we extend entropy and its relevance in XXXXX , e.g . microsate , ensemble , and information , to evaluate the XXXXX of elastic bodies , which are discretized with finite elements , some concepts in contact mechanics are considered as the same or similar meaning of that in XXXXX', ['statistical physics', 'contact forces'])\n", "('method used for task', 'we extend entropy and its relevance in XXXXX , e.g . microsate , ensemble , and information , to evaluate the contact forces of elastic bodies , which are discretized with XXXXX , some concepts in contact mechanics are considered as the same or similar meaning of that in XXXXX', ['statistical physics', 'finite elements'])\n", "('NONE', 'we extend entropy and its relevance in statistical physics , e.g . microsate , ensemble , and information , to evaluate the XXXXX of elastic bodies , which are discretized with XXXXX , some concepts in contact mechanics are considered as the same or similar meaning of that in statistical physics', ['contact forces', 'finite elements'])\n", "('method used for task', 'we extend entropy and its relevance in XXXXX , e.g . microsate , ensemble , and information , to evaluate the XXXXX of elastic bodies , which are discretized with finite elements , some concepts in contact mechanics are considered as the same or similar meaning of that in XXXXX', ['contact forces', 'statistical physics'])\n", "('method used for task', 'we extend entropy and its relevance in XXXXX , e.g . microsate , ensemble , and information , to evaluate the contact forces of elastic bodies , which are discretized with XXXXX , some concepts in contact mechanics are considered as the same or similar meaning of that in XXXXX', ['finite elements', 'statistical physics'])\n", "('method used for task', 'we obtain an explicit XXXXX for the normalized XXXXX by maximizing entropy subject to an expectation value and the principle of minimum potential energy', ['probability distribution', 'contact forces'])\n", "('method used for task', 'we obtain an explicit XXXXX for the normalized contact forces by maximizing entropy subject to an expectation value and the principle of minimum XXXXX', ['probability distribution', 'potential energy'])\n", "('method used for task', 'we obtain an explicit probability distribution for the normalized XXXXX by maximizing entropy subject to an expectation value and the principle of minimum XXXXX', ['contact forces', 'potential energy'])\n", "('NONE', 'we construct an efficient XXXXX by solving a series of isolated systems to obtain the XXXXX , and give a novelty termination criteria to terminate the iteration', ['iterative procedure', 'contact forces'])\n", "('NONE', 'implemented two-scale XXXXX as a XXXXX for a 3d an eight-node interface element', ['damage model', 'material model'])\n", "('NONE', 'we present a new approach to compute the XXXXX of solid XXXXX', ['mass matrix', 'finite elements'])\n", "('NONE', 'XXXXX focus on consistent XXXXX of 10-node tetrahedral element', ['numerical examples', 'mass matrix'])\n", "('NONE', 'we specify an approach for the XXXXX of XXXXX', ['dynamic reconfiguration', 'mobile applications'])\n", "('method used for task', 'our offline learning approach with an XXXXX outperforms existing methods for XXXXX', ['adaptive genetic algorithm', 'model calibration'])\n", "('method used for task', 'employ hybrid XXXXX to generate representative views for XXXXX', ['3d models', 'feature lines'])\n", "('method used for task', 'we propose a XXXXX to efficiently solve the XXXXX', ['iterative algorithm', 'non-linear model'])\n", "('NONE', 'we define the well-posedness of XXXXX of the rational XXXXX', ['control polygon', 'bezier curve'])\n", "('method used for task', 'the XXXXX is injective if and only if its XXXXX is well-posed', ['bezier curve', 'control polygon'])\n", "('NONE', 'we present a XXXXX to determine the injectivity of the XXXXX', ['geometric method', 'bezier curve'])\n", "('NONE', 'geometrically XXXXX can be recognized by anatomists as meaningful locations on 3d XXXXX scans', ['salient points', 'human body'])\n", "('method used for task', 'efficient XXXXX for XXXXX is presented', ['compression algorithm', 'triangle meshes'])\n", "('method used for task', 'a method for identifying potential XXXXX that are subordinate to different XXXXX is present', ['feature points', 'feature lines'])\n", "('NONE', 'a feature selection framework is proposed to achieve XXXXX model-free XXXXX', ['high performance', 'gait recognition'])\n", "('method used for task', 'a XXXXX is proposed to achieve XXXXX model-free gait recognition', ['high performance', 'feature selection framework'])\n", "('method used for task', 'a XXXXX is proposed to achieve high performance model-free XXXXX', ['gait recognition', 'feature selection framework'])\n", "('method used for task', 'novel XXXXX based visual ego-motion estimation algorithm robust to abrupt XXXXX', ['particle filtering', 'camera motion'])\n", "('method used for task', 'multi-kernel XXXXX is a new facial XXXXX', ['appearance model', 'point detector'])\n", "('NONE', 'the XXXXX of the simplest XXXXX is barely increased', ['computational cost', 'modeling strategy'])\n", "('method used for task', 'introducing a XXXXX for simulating XXXXX', ['stochastic model', 'eye movements'])\n", "('method used for task', 'it extends the XXXXX to incorporate color and XXXXX', ['fast marching method', 'depth information'])\n", "('NONE', 'no strong constraints are imposed on the XXXXX or the XXXXX', ['camera motion', 'target appearance'])\n", "('method used for task', 'a XXXXX ( gp ) approach is proposed to design a combined XXXXX for image segmentation algorithms', ['genetic programming', 'evaluation measure'])\n", "('method used for task', 'a XXXXX ( gp ) approach is proposed to design a combined evaluation measure for XXXXX', ['genetic programming', 'image segmentation algorithms'])\n", "('method used for task', 'a genetic programming ( gp ) approach is proposed to design a combined XXXXX for XXXXX', ['evaluation measure', 'image segmentation algorithms'])\n", "('method used for task', 'we proposed a novel XXXXX based model for inhomogeneous XXXXX', ['level set', 'image segmentation'])\n", "('NONE', 'the local XXXXX can significantly increase XXXXX', ['image information', 'image contrast'])\n", "('method used for task', 'examined nine well-known models of XXXXX using the best metric to identify the best XXXXX models', ['visual saliency', 'visual saliency models'])\n", "('NONE', 'use of novel XXXXX with stereo features and XXXXX to disparity', ['appearance model', 'plane fitting'])\n", "('NONE', 'facade segmentation by XXXXX : hierarchical XXXXX', ['graphical model', 'markov random field'])\n", "('method used for task', 'allows the accurate regression of XXXXX to be scaled to XXXXX', ['gaussian processes', 'large data'])\n", "('NONE', 'we propose two strategies for XXXXX through XXXXX', ['face recognition', 'multiple features'])\n", "('method used for task', 'the XXXXX and XXXXX are learned simultaneously', ['dictionary learning', 'fusion process'])\n", "('method used for task', 'any XXXXX can be used , and errors are calculated on the original XXXXX', ['camera model', 'image space'])\n", "('NONE', 'the estimation starts with a XXXXX and ends with a XXXXX', ['random search', 'continuous optimization'])\n", "('NONE', 'XXXXX taking into account the XXXXX is introduced', ['distance function', 'hierarchical structure'])\n", "('method used for task', 'we propose an XXXXX for tracking XXXXX , capable of handling anomalies', ['online method', 'dense crowds'])\n", "('NONE', 'we propose a XXXXX of the problem based on XXXXX', ['nonlinear optimization', 'mutual information'])\n", "('NONE', 'we use XXXXX to find the scene that best renders the XXXXX', ['nonlinear optimization', 'input image'])\n", "('NONE', 'three XXXXX over two XXXXX are employed to support the conclusions', ['statistical tests', 'performance measures'])\n", "('NONE', 'XXXXX was implemented using bayes XXXXX', ['model comparison', 'information criterion'])\n", "('method used for task', 'XXXXX is incorporated by employing XXXXX into clips', ['motion analysis', 'motion segmentation'])\n", "('method used for task', 'we propose an ensemble XXXXX ( edl ) framework for XXXXX', ['dictionary learning', 'saliency detection'])\n", "('method used for task', 'XXXXX based method for XXXXX using multiple views', ['data fusion', 'activity recognition'])\n", "('method used for task', 'XXXXX based method for activity recognition using XXXXX', ['data fusion', 'multiple views'])\n", "('NONE', 'data fusion based method for XXXXX using XXXXX', ['activity recognition', 'multiple views'])\n", "('NONE', 'we evaluate the XXXXX in XXXXX', ['object recognition', '3d shape descriptor'])\n", "('NONE', 'eyebrow gestures and periodic XXXXX convey XXXXX', ['linguistic information', 'head movements'])\n", "('NONE', 'our methods can enhance XXXXX lying in the low frequency part of XXXXX', ['face features', 'face images'])\n", "('NONE', 'a statistical XXXXX yields the complexity of the XXXXX', ['stability analysis', 'registration model'])\n", "('NONE', 'we reduce the XXXXX of fingertip locations conditioned to a XXXXX', ['search space', 'hand gesture'])\n", "('method used for task', 'XXXXX of 99 % ( almost perfect ) is reached to detect XXXXX', ['classification accuracy', 'mask attacks'])\n", "('method used for task', 'XXXXX tackling speaker dependency , head poses and XXXXX', ['visual features', 'temporal information'])\n", "('NONE', 'for XXXXX , a hierarchical piece-wise XXXXX is examined', ['non-rigid registration', 'affine transform'])\n", "('method used for task', \"we examine XXXXX on the proposed metric and moran 's XXXXX\", ['roc analysis', 'spatial correlation'])\n", "('method used for task', 'XXXXX is used to combine various XXXXX and determine a safety score', ['fuzzy logic', 'attribute values'])\n", "('NONE', 'XXXXX provides a XXXXX for textural images in each class', ['sparse representation', 'compact model'])\n", "('NONE', 'distributions of gmrf local XXXXX as improved XXXXX', ['texture descriptors', 'parameter estimates'])\n", "('method used for task', 'we propose a novel XXXXX for XXXXX', ['action recognition', 'deep learning model'])\n", "('method used for task', 'an integration of XXXXX , machine listening , and XXXXX', ['computer vision', 'machine learning'])\n", "('NONE', 'we introduce the 3dspmk for object and XXXXX in XXXXX', ['scene recognition', 'depth images'])\n", "('method used for task', 'we design a scale-invariant XXXXX for shape matching and XXXXX', ['shape descriptor', 'object detection'])\n", "('NONE', 'the proposed tracker combines the flexibility of XXXXX and robustness of XXXXX', ['interest points', 'sparse representation'])\n", "('method used for task', 'a directed XXXXX is proposed for XXXXX', ['structural model', 'feature correspondence'])\n", "('method used for task', 'we propose two novel XXXXX for unsupervised non-gaussian XXXXX', ['feature selection', 'inference frameworks'])\n", "('method used for task', 'a regional bounding spherical descriptor is used for XXXXX and XXXXX', ['3d face', 'emotion analysis'])\n", "('method used for task', 'dst-klpp provides higher XXXXX than other methods , including wavelet , curvelet and XXXXX', ['contourlet transform', 'classification rates'])\n", "('method used for task', 'the proposed model gives a XXXXX and a good balance between XXXXX and processing efficiency', ['real-time performance', 'segmentation accuracy'])\n", "('method used for task', 'a study of the use of XXXXX for XXXXX is presented', ['topic models', 'video retrieval'])\n", "('NONE', 'the results highlight the XXXXX among the XXXXX', ['topic models', 'performance differences'])\n", "('method used for task', 'a XXXXX has been introduced in the process to reduce the XXXXX of particle filtering', ['markov chain model', 'computational complexity'])\n", "('NONE', 'a XXXXX has been introduced in the process to reduce the computational complexity of XXXXX', ['markov chain model', 'particle filtering'])\n", "('NONE', 'a markov chain model has been introduced in the process to reduce the XXXXX of XXXXX', ['computational complexity', 'particle filtering'])\n", "('NONE', 'application of XXXXX along with a XXXXX to refine any outliers', ['ransac algorithm', 'histogram technique'])\n", "('method used for task', 'hierarchical XXXXX is proposed to obtain an acceptable solution in XXXXX', ['real time', 'diffusion method'])\n", "('method used for task', 'XXXXX and XXXXX between objects and shadows are used to recover misdetected shadows', ['mutual information', 'data association'])\n", "('method used for task', 'a effective XXXXX is introduced to optimize the XXXXX', ['iterative algorithm', 'objective function'])\n", "('NONE', 'new multi-lateral filter to efficiently increase the XXXXX of low-resolution and noisy XXXXX in real-time', ['spatial resolution', 'depth maps'])\n", "('NONE', 'the proposed filter has been effectively and efficiently implemented for XXXXX in XXXXX', ['dynamic scenes', 'real-time applications'])\n", "('NONE', 'we examine the role of XXXXX and image semantics in understanding XXXXX', ['visual attention', 'image memorability'])\n", "('method used for task', 'we propose an attention-driven spatial XXXXX for XXXXX', ['pooling strategy', 'image memorability'])\n", "('NONE', 'an algorithm has been developed for XXXXX of XXXXX', ['automatic detection', 'vanishing points'])\n", "('method used for task', 'the results have been analyzed according to the number of XXXXX and XXXXX', ['vanishing points', 'image resolution'])\n", "('method used for task', 'XXXXX is done by the pre-trained svm using 7 different XXXXX', ['state transition', 'input features'])\n", "('method used for task', 'we also learn XXXXX from very few positive and related XXXXX', ['event detectors', 'training samples'])\n", "('method used for task', 'in the XXXXX each region gives rise to a learnt weak XXXXX', ['ranking function', 'training step'])\n", "('method used for task', 'proposed XXXXX for XXXXX', ['ensemble learning', 'face hallucination'])\n", "('method used for task', 'designed a pre-evaluation stage to save XXXXX and reduce XXXXX', ['computation time', 'false alarm rate'])\n", "('method used for task', 'symstereo and pearl are combined for carrying the XXXXX and XXXXX simultaneously', ['3d reconstruction', 'plane fitting'])\n", "('NONE', 'a new dual many-to-one encoder method for XXXXX across XXXXX', ['feature extraction', 'action datasets'])\n", "('method used for task', 'no training or XXXXX is needed and achieve XXXXX', ['camera calibration', 'real-time processing'])\n", "('method used for task', 'we propose a new local XXXXX for XXXXX', ['action recognition', 'part model'])\n", "('method used for task', 'XXXXX and fast XXXXX are achieved', ['high performance', 'action recognition'])\n", "('NONE', 'a 3-axis gyro mounted to a XXXXX can frequently predict the main component of XXXXX', ['video camera', 'optical flow'])\n", "('method used for task', 'gyro regularization adds very little XXXXX to XXXXX', ['computational cost', 'feature tracking'])\n", "('NONE', 'the new XXXXX improves results in complex XXXXX', ['kernel framework', 'activities recognition'])\n", "('method used for task', 'we apply collective XXXXX to cross-domain XXXXX', ['matrix factorization', 'action recognition'])\n", "('NONE', 'XXXXX learned by the structured output XXXXX', ['support vector machines', 'landmark detector'])\n", "('method used for task', 'we simultaneously learn XXXXX and XXXXX', ['visual features', 'hash functions'])\n", "('method used for task', 'a timely review of XXXXX based on spatio-temporal XXXXX', ['action recognition', 'local features'])\n", "('method used for task', 'transfer techniques in XXXXX to video domain for XXXXX', ['action recognition', 'image domain'])\n", "('method used for task', 'handle the misalignment problems in XXXXX and XXXXX', ['image classification', 'action recognition'])\n", "('NONE', 'improve the discrimination of XXXXX in XXXXX', ['feature pooling', 'visual recognition tasks'])\n", "('NONE', 'uses the notion of XXXXX and a pattern comic to make sense of the constitutive ‘mechanics’ of XXXXX', ['data structures', 'business patterns'])\n", "('NONE', 'grounds the application of XXXXX and pattern comics in material collected within an XXXXX study', ['action research', 'business patterns'])\n", "('method used for task', 'we propose a link-bridged XXXXX for cross-domain XXXXX', ['topic model', 'document classification'])\n", "('NONE', 'directions for future works related to XXXXX in XXXXX', ['metadata quality', 'data infrastructures'])\n", "('NONE', 'we present an XXXXX for the study of cross-lingual XXXXX', ['evaluation framework', 'link discovery'])\n", "('NONE', 'XXXXX contain XXXXX and evolve over time', ['xml documents', 'temporal information'])\n", "('NONE', 'our work explores XXXXX , versioning and querying support in XXXXX', ['change detection', 'xml documents'])\n", "('NONE', 'we investigate various query XXXXX using XXXXX', ['estimation methods', 'bias–variance analysis'])\n", "('NONE', 'we propose a new method for XXXXX validity using cross XXXXX', ['outlier detection', 'term translation'])\n", "('method used for task', 'we show that variance of rn sampling grows with XXXXX for XXXXX', ['scale-free networks', 'data size'])\n", "('NONE', 'we model bid XXXXX as a mixed XXXXX', ['keyword suggestion', 'integer optimization problem'])\n", "('NONE', 'enriching queries using XXXXX increases XXXXX in healthcare domain', ['user preferences', 'information retrieval'])\n", "('NONE', 'prioritizing XXXXX improves XXXXX', ['user preferences', 'service discovery'])\n", "('method used for task', 'we combine active and XXXXX for cross-lingual XXXXX', ['semi-supervised learning', 'sentiment classification'])\n", "('method used for task', 'density analysis of XXXXX is used in XXXXX', ['unlabeled data', 'active learning'])\n", "('NONE', 'XXXXX of XXXXX is used in active learning', ['unlabeled data', 'density analysis'])\n", "('NONE', 'XXXXX of unlabeled data is used in XXXXX', ['active learning', 'density analysis'])\n", "('NONE', 'results show that incorporating XXXXX can speed up XXXXX', ['learning process', 'density analysis'])\n", "('NONE', 'we implement an active learning scenario for XXXXX on XXXXX', ['sentiment analysis', 'data streams'])\n", "('method used for task', 'XXXXX are shown to be suitable for XXXXX', ['sentiment analysis', 'machine learning methods'])\n", "('NONE', 'XXXXX improves the accuracy of XXXXX', ['active learning', 'sentiment classification'])\n", "('method used for task', 'the use of wikipedia as a XXXXX for XXXXX', ['knowledge source', 'question answering system'])\n", "('NONE', 'experimental results indicate XXXXX in term of XXXXX', ['high accuracy', 'bleu score'])\n", "('NONE', 'we explore state-of-the-art supervised machine learning methods for XXXXX of czech XXXXX', ['sentiment analysis', 'social media'])\n", "('method used for task', 'we explore state-of-the-art XXXXX for XXXXX of czech social media', ['sentiment analysis', 'supervised machine learning methods'])\n", "('NONE', 'we explore state-of-the-art XXXXX for sentiment analysis of czech XXXXX', ['social media', 'supervised machine learning methods'])\n", "('method used for task', 'combine grouping of video search results with XXXXX to assist XXXXX', ['video retrieval', 'recommendation techniques'])\n", "('NONE', 'we carry out an XXXXX to determine characteristics of XXXXX', ['empirical analysis', 'social media channels'])\n", "('method used for task', 'a new XXXXX for the next-generation XXXXX', ['theoretical framework', 'recommender systems'])\n", "('NONE', 'XXXXX of learning couplings in XXXXX and recommendation', ['case studies', 'data mining'])\n", "('NONE', 'the rise of XXXXX has fueled interest in XXXXX', ['social media', 'sentiment classification'])\n", "('method used for task', 'relationship between XXXXX and XXXXX : common and different reasons among different knowledge groups', ['task difficulty', 'user knowledge'])\n", "('NONE', 'we tackle the problem of XXXXX differently , by XXXXX', ['data fusion', 'microblog search'])\n", "('NONE', 'tuning XXXXX increases XXXXX', ['genetic algorithm', 'time efficiency'])\n", "('NONE', 'expressive signals enrich the XXXXX of baseline and XXXXX', ['feature space', 'ensemble classifiers'])\n", "('method used for task', 'means uses XXXXX and standards for XXXXX and integration', ['semantic web technologies', 'data sharing'])\n", "('NONE', 'we carry out an XXXXX to determine characteristics of XXXXX', ['empirical analysis', 'social media channels'])\n", "('NONE', 'we explore state-of-the-art supervised machine learning methods for XXXXX of czech XXXXX', ['sentiment analysis', 'social media'])\n", "('method used for task', 'we explore state-of-the-art XXXXX for XXXXX of czech social media', ['sentiment analysis', 'supervised machine learning methods'])\n", "('NONE', 'we explore state-of-the-art XXXXX for sentiment analysis of czech XXXXX', ['social media', 'supervised machine learning methods'])\n", "('NONE', 'we face the XXXXX of having a limited set of XXXXX', ['pairwise constraints', 'real-world problem'])\n", "('method used for task', 'show effectiveness with above 70 % XXXXX for user XXXXX', ['prediction accuracy', 'search performance'])\n", "('method used for task', 'XXXXX effectively employed to improve XXXXX', ['unlabeled data', 'classification performance'])\n", "('NONE', 'we build a XXXXX that can be tested by the XXXXX', ['prototype system', 'research community'])\n", "('NONE', 'improving XXXXX by integrating co-occurrence relations into XXXXX', ['translation quality', 'word models'])\n", "('NONE', 'comparing different estimations of XXXXX from XXXXX', ['translation probabilities', 'word correlations'])\n", "('NONE', 'we perform XXXXX of personality , XXXXX and mood sharing in twitter', ['correlation analysis', 'communication style'])\n", "('method used for task', 'XXXXX ( gp ) algorithm has been employed for XXXXX', ['genetic programming', 'feature learning'])\n", "('method used for task', 'we present a new XXXXX for relations between concepts based on XXXXX of concepts', ['weighting scheme', 'distributed representations'])\n", "('NONE', 'examine the XXXXX of XXXXX on gens y and z', ['the internet', 'influence factors'])\n", "('NONE', 'we study the characteristics of the XXXXX ( sk ) of XXXXX', ['straight skeleton', 'monotone polygons'])\n", "('method used for task', 'we propose a XXXXX and a XXXXX for social collaboration processes', ['modeling approach', 'visual modeling notation'])\n", "('NONE', 'framework of ten XXXXX functionalities ( cmf ) in XXXXX is defined', ['business processes', 'compliance monitoring'])\n", "('method used for task', 'XXXXX and presentation of real-world business constraints for compliance are extracted from five XXXXX', ['systematic literature review', 'case studies'])\n", "('NONE', 'we consider the comparison with rtl power estimation techniques on several design aspects : XXXXX , XXXXX and modeling effort ,', ['estimation accuracy', 'simulation time'])\n", "('NONE', 'XXXXX appropriate for predictive maintenance of XXXXX', ['mathematical model', 'transmission line'])\n", "('NONE', 'distributed XXXXX and improvement of delivered XXXXX', ['power quality', 'energy resources'])\n", "('NONE', 'examines a XXXXX of normative prototypes that reshape the issue of XXXXX', ['case study', 'aircraft noise'])\n", "('NONE', 'XXXXX avoids XXXXX in ihe xds shared ehr systems', ['content-based search', 'information overload'])\n", "('method used for task', 'XXXXX can be used to structure information from XXXXX reports', ['natural language processing', 'free text'])\n", "('method used for task', 'XXXXX can be used to XXXXX from free text reports', ['natural language processing', 'structure information'])\n", "('NONE', 'natural language processing can be used to XXXXX from XXXXX reports', ['free text', 'structure information'])\n", "('method used for task', 'XXXXX approaches are yet to be fully explored in XXXXX of cancer information', ['machine learning', 'text mining'])\n", "('method used for task', 'XXXXX are feasible platforms for self-supervised XXXXX', ['mobile devices', 'cognitive training'])\n", "('method used for task', 'self-guided and internet-delivered cognitive XXXXX ( icbt ) has potential for XXXXX', ['older adults', 'behavior therapy'])\n", "('NONE', 'the study result show that adults with adhd benefit from a coach-guided XXXXX using smartphone applications and that XXXXX are a feasible way to reach this patient group', ['internet interventions', 'online intervention'])\n", "('NONE', 'the study result show that adults with adhd benefit from a coach-guided online intervention using XXXXX and that XXXXX are a feasible way to reach this patient group', ['internet interventions', 'smartphone applications'])\n", "('NONE', 'the study result show that adults with adhd benefit from a coach-guided XXXXX using XXXXX and that internet interventions are a feasible way to reach this patient group', ['online intervention', 'smartphone applications'])\n", "('method used for task', 'the effectiveness of a guided and unguided act-based XXXXX for XXXXX ( actonpain ) will be investigated', ['chronic pain', 'online intervention'])\n", "('NONE', 'cognitive behavioural therapy via XXXXX ( icbt ) may benefit XXXXX', ['the internet', 'university students'])\n", "('NONE', 'a XXXXX of client e-mails was conducted within the setting of a guided internet-based cognitive XXXXX program for depression', ['content analysis', 'behavior therapy'])\n", "('method used for task', 'XXXXX is a depleting XXXXX , however the availability of effective treatments are scarce', ['chronic pain', 'health problem'])\n", "('method used for task', 'the presented method provides a XXXXX as basis for XXXXX', ['bayesian framework', 'data fusion'])\n", "('method used for task', 'XXXXX is interpreted by a XXXXX for copd-exacerbation detection', ['patient data', 'bayesian network model'])\n", "('method used for task', 'explorer provides high XXXXX and strong XXXXX', ['estimation accuracy', 'privacy protection'])\n", "('NONE', 'automatic pgx-specific drug–gene XXXXX from XXXXX is difficult', ['free text', 'relationship extraction'])\n", "('NONE', 'we develop a semi-supervised XXXXX method requiring minimal prior XXXXX', ['domain knowledge', 'relationship extraction'])\n", "('method used for task', 'we created software integrated model for XXXXX and XXXXX ( impact )', ['patient care', 'clinical trials'])\n", "('NONE', 'impact coordinates XXXXX visits with XXXXX visits', ['clinical research', 'patient care'])\n", "('NONE', 'examines health outcomes , effects and affordances of XXXXX in XXXXX', ['social media', 'chronic disease'])\n", "('method used for task', 'XXXXX is needed in XXXXX of gene expression', ['gene selection', 'supervised classification'])\n", "('NONE', 'XXXXX is needed in supervised classification of XXXXX', ['gene selection', 'gene expression'])\n", "('NONE', 'gene selection is needed in XXXXX of XXXXX', ['supervised classification', 'gene expression'])\n", "('NONE', 'XXXXX of electronic XXXXX is achieved for safety studies', ['secondary use', 'health records'])\n", "('method used for task', 'XXXXX of electronic health records is achieved for XXXXX', ['secondary use', 'safety studies'])\n", "('method used for task', 'secondary use of electronic XXXXX is achieved for XXXXX', ['health records', 'safety studies'])\n", "('method used for task', 'combining XXXXX and XXXXX , consensus will be fostered', ['clinical decision support', 'social network'])\n", "('NONE', 'XXXXX with the XXXXX encourages adoption and collaboration', ['open source', 'social network'])\n", "('method used for task', 'we review the use of XXXXX for health applied to XXXXX', ['mobile phones', 'older adults'])\n", "('NONE', 'the XXXXX was tested on two XXXXX using 2150 production radiology reports', ['similarity measure', 'classification problems'])\n", "('NONE', 'XXXXX can exploit the vast amounts of XXXXX stored in emrs', ['semi-supervised learning', 'unlabeled data'])\n", "('method used for task', 'XXXXX and XXXXX technologies can be used to deal with biology’s XXXXX sets', ['cloud computing', 'big data'])\n", "('method used for task', 'XXXXX and XXXXX can be used to deal with biology’s big data sets', ['cloud computing', 'big data technologies'])\n", "('method used for task', 'XXXXX and big data technologies can be used to deal with biology’s XXXXX', ['cloud computing', 'big data sets'])\n", "('method used for task', 'cloud computing and XXXXX technologies can be used to deal with biology’s XXXXX sets', ['big data', 'big data technologies'])\n", "('method used for task', 'cloud computing and XXXXX technologies can be used to deal with biology’s XXXXX sets', ['big data', 'big data sets'])\n", "('method used for task', 'cloud computing and XXXXX can be used to deal with biology’s XXXXX', ['big data technologies', 'big data sets'])\n", "('method used for task', 'challenges associated with XXXXX and XXXXX in biology are discussed', ['cloud computing', 'big data technologies'])\n", "('NONE', 'three different models for inferring XXXXX from XXXXX are proposed', ['gene networks', 'microarray data'])\n", "('NONE', 'semantator converts biomedical text to XXXXX with a XXXXX', ['linked data', 'formal semantics'])\n", "('method used for task', 'we develop a new XXXXX and query-document XXXXX', ['vector space model', 'similarity function'])\n", "('NONE', 'successfully used as XXXXX in the semeval-2013 ddi XXXXX', ['gold standard', 'extraction task'])\n", "('method used for task', 'a dynamic tag cloud may reduce XXXXX for XXXXX', ['information overload', 'clinical trial search'])\n", "('method used for task', 'we investigated how XXXXX and XXXXX changed over time', ['user interaction', 'usability problems'])\n", "('method used for task', 'we propose a XXXXX to discover underlying patterns in XXXXX', ['probabilistic model', 'clinical pathways'])\n", "('NONE', 'XXXXX of XXXXX and mappings evolution', ['quantitative analysis', 'medical ontologies'])\n", "('method used for task', 'we proposed a XXXXX ( graph-based and machine-learning ) for temporal XXXXX', ['hybrid system', 'relation extraction'])\n", "('NONE', 'XXXXX shows superiority of our XXXXX on our utility measures', ['empirical evaluation', 'model based'])\n", "('NONE', 'XXXXX can support XXXXX and analysis', ['classification systems', 'data retrieval'])\n", "('NONE', 'discovery of XXXXX in XXXXX using literature knowledge', ['drug–drug interactions', 'patient data'])\n", "('NONE', 'potential XXXXX identified in medication lists from XXXXX', ['drug–drug interactions', 'clinical data'])\n", "('method used for task', 'XXXXX with globus transfer for high-performance and reliable XXXXX', ['data transfer', 'integrate galaxy'])\n", "('method used for task', 'XXXXX with htcondor scheduler for auto-scaling and XXXXX', ['parallel computing', 'integrate galaxy'])\n", "('method used for task', 'two bioinformatics workflow XXXXX and XXXXX are presented', ['use cases', 'performance evaluation'])\n", "('NONE', 'our approach combines XXXXX visual queries , mining , and XXXXX', ['ad hoc', 'interactive visualization'])\n", "('NONE', 'a feature-based approach identifies XXXXX with similar XXXXX', ['clinical trials', 'eligibility text'])\n", "('method used for task', 'we propose a XXXXX for detecting XXXXX in mammograms', ['cad system', 'breast cancer'])\n", "('NONE', 'XXXXX optimized XXXXX detects the cancers', ['swarm intelligence', 'wavelet neural network'])\n", "('method used for task', 'we focus on optimized XXXXX to enhance the XXXXX', ['wavelet neural network', 'detection accuracy'])\n", "('NONE', 'disorders , findings , drugs and XXXXX were annotated in swedish XXXXX', ['clinical text', 'body parts'])\n", "('method used for task', 'we propose a new prognostic XXXXX based on a bayesian XXXXX', ['prediction model', 'evolutionary learning'])\n", "('NONE', 'the proposed bayesian XXXXX effectively handles a complex XXXXX', ['evolutionary learning', 'search space'])\n", "('NONE', 'the performance of XXXXX outperforms other XXXXX', ['classification models', 'hypergraph classifiers'])\n", "('NONE', 'XXXXX of XXXXX for research is a complex task but it benefits the entire medical center community', ['secondary use', 'clinical data'])\n", "('method used for task', 'change in XXXXX and XXXXX were integrated simultaneously', ['gene expression', 'structural information'])\n", "('NONE', 'XXXXX include XXXXX , task accuracy , and path length', ['performance metrics', 'completion time'])\n", "('method used for task', 'XXXXX include completion time , task accuracy , and XXXXX', ['performance metrics', 'path length'])\n", "('method used for task', 'performance metrics include XXXXX , task accuracy , and XXXXX', ['completion time', 'path length'])\n", "('NONE', 'how XXXXX varies with accuracy index and XXXXX', ['sample size', 'effect size'])\n", "('method used for task', 'XXXXX is gathered by a wearable sensor and sent to a XXXXX', ['mobile device', 'ecg data'])\n", "('NONE', 'we propose an XXXXX that considers the semantic of XXXXX', ['medical images', 'image retrieval system'])\n", "('method used for task', 'XXXXX is used to search for a good XXXXX', ['evolutionary programming', 'discretization scheme'])\n", "('NONE', 'XXXXX observed in colposcopy are used as predictors of XXXXX', ['temporal patterns', 'cervical cancer'])\n", "('NONE', 'propose XXXXX that allow XXXXX to make collective decisions', ['ensemble methods', 'base classifiers'])\n", "('NONE', 'two strategies enhanced the XXXXX ( dt ) proteins prediction power of XXXXX', ['drug target', 'svm classifier'])\n", "('method used for task', 'demonstrated utility of an in-domain collection ( XXXXX ) for XXXXX', ['clinical text', 'query expansion'])\n", "('NONE', 'the release of electronic XXXXX in the form of XXXXX may lead to privacy violations', ['health data', 'data streams'])\n", "('method used for task', 'XXXXX are shaped by the healthcare process and patient XXXXX', ['measurement patterns', 'health states'])\n", "('method used for task', 'XXXXX : the probability for suffering XXXXX in encrypted form', ['case study', 'cardiovascular disease'])\n", "('NONE', 'challenges include XXXXX , access , and limits of XXXXX and technology', ['data quality', 'human cognition'])\n", "('method used for task', 'we analyzed XXXXX and latent XXXXX for agreement on ctg evaluations', ['majority voting', 'class model'])\n", "('NONE', 'we built a novel architecture for XXXXX of XXXXX', ['information retrieval', 'patient records'])\n", "('NONE', 'annotating raw XXXXX generated the highest XXXXX', ['clinical text', 'quality data'])\n", "('NONE', 'XXXXX in electronic XXXXX ( ehr ) may re-identify patients', ['diagnosis codes', 'health records'])\n", "('NONE', 'we study discrete medical XXXXX with XXXXX', ['multi-label classification', 'time series datasets'])\n", "('method used for task', 'we present 3dfd.ujaen.es , a XXXXX for computing and analyzing the 3d XXXXX ( 3dfd )', ['web platform', 'fractal dimension'])\n", "('method used for task', 'the first XXXXX that allows the users to calculate , visualize , analyze and compare the 3dfd from XXXXX ( i.e . mri )', ['web platform', '3d images'])\n", "('NONE', 'we measure cds impact on XXXXX using statistical and XXXXX', ['cancer screening', 'simulation models'])\n", "('NONE', 'systematic studies of drug side XXXXX can facilitate XXXXX', ['drug discovery', 'effect associations'])\n", "('NONE', 'there is a need for techniques for XXXXX in medical XXXXX with events', ['data mining', 'time series'])\n", "('method used for task', 'XXXXX are automatically generated based on a XXXXX', ['regular expressions', 'domain ontology'])\n", "('NONE', 'first XXXXX of XXXXX in the biomedical domain', ['systematic review', 'text summarization'])\n", "('NONE', 'resolving the complexity of XXXXX from entity–attribute–value XXXXX', ['data extraction', 'data model'])\n", "('NONE', 'XXXXX may help biomedical researchers manage XXXXX', ['automatic summarization', 'information overload'])\n", "('NONE', 'ontological integration of XXXXX elements can compensate for deficiencies in XXXXX', ['data quality', 'ehr data'])\n", "('method used for task', 'the proposed XXXXX is achieved in the link layer without XXXXX', ['routing algorithm', 'route discovery'])\n", "('NONE', 'riig facilitates finding effective XXXXX in a XXXXX setting', ['drug targets', 'personalized medicine'])\n", "('NONE', 'riig bridges curated biomedical knowledge and XXXXX with XXXXX', ['causal reasoning', 'clinical data'])\n", "('NONE', 'we found evidence that the emr XXXXX are outpacing XXXXX', ['adoption changes', 'informatics research'])\n", "('method used for task', 'we propose an effective XXXXX for inferring XXXXX with priors', ['hybrid method', 'bayesian networks'])\n", "('NONE', 'the multi-technique approach includes XXXXX , XXXXX and ontology-based knowledgebase', ['information retrieval', 'natural language processing'])\n", "('NONE', 'two XXXXX are integrated in saps : XXXXX and pattern search', ['optimization methods', 'simulated annealing'])\n", "('method used for task', 'two XXXXX are integrated in saps : simulated annealing and XXXXX', ['optimization methods', 'pattern search'])\n", "('method used for task', 'two optimization methods are integrated in saps : XXXXX and XXXXX', ['simulated annealing', 'pattern search'])\n", "('NONE', 'XXXXX , such as visibility and XXXXX , is taken into account', ['prior knowledge', 'spatial constraints'])\n", "('method used for task', 'XXXXX ( color ) and XXXXX ( radius ) are depicted in circles', ['effect size', 'statistical significance'])\n", "('method used for task', 'we used XXXXX to model relations between outbreak and algorithm characteristics and XXXXX', ['bayesian networks', 'detection performance'])\n", "('NONE', 'we show that using nlp-based XXXXX improves adr XXXXX', ['feature extraction', 'classification accuracy'])\n", "('method used for task', 'a compilation of useful XXXXX for XXXXX is presented', ['bioinformatics tools', 'vaccine development'])\n", "('method used for task', 'we devise a XXXXX based algorithm on the reliable XXXXX', ['random walk', 'heterogeneous network'])\n", "('NONE', 'we discuss methods to use XXXXX in XXXXX for predictive modeling', ['temporal information', 'ehr data'])\n", "('NONE', 'our method was able to use XXXXX in XXXXX to improve performance', ['temporal information', 'ehr data'])\n", "('method used for task', 'XXXXX was used to estimate XXXXX of predictive features', ['multiple imputation', 'missing values'])\n", "('method used for task', 'XXXXX and XXXXX exploit information extrapolated from the unified medical language system', ['semantic similarity', 'relatedness measures'])\n", "('method used for task', 'evaluates combining XXXXX and XXXXX', ['semantic similarity', 'relatedness measures'])\n", "('NONE', 'we describe a method to generate word-concept XXXXX from a XXXXX', ['statistical models', 'knowledge base'])\n", "('method used for task', 'usage of XXXXX to create a XXXXX of the therapy process', ['hidden markov models', 'stochastic model'])\n", "('method used for task', 'the design and use of a XXXXX meter is analyzed using XXXXX', ['blood glucose', 'distributed cognition'])\n", "('NONE', 'we construct multiple XXXXX to make full use of all classes of XXXXX not only target samples', ['svdd models', 'training samples'])\n", "('method used for task', 'new method to determine XXXXX for building logistic XXXXX', ['sample size', 'prediction model'])\n", "('method used for task', 'XXXXX and aggregation techniques facilitate XXXXX', ['semantic similarity', 'hierarchical clustering'])\n", "('NONE', 'we constructed disease-related XXXXX using literature and XXXXX', ['gene network', 'google data'])\n", "('NONE', 'improved performance of the prediction system is reported using the XXXXX obtained through XXXXX', ['optimal threshold', 'pso algorithm'])\n", "('NONE', 'the extracted XXXXX were visualized as XXXXX and pathways', ['ppi information', 'interaction networks'])\n", "('NONE', 'we characterized the pathway-level XXXXX of XXXXX', ['topological properties', 'drug targets'])\n", "('method used for task', 'we optimized XXXXX based on pathway-level XXXXX', ['drug targets', 'topological properties'])\n", "('method used for task', 'XXXXX and XXXXX perform best in terms of auc', ['random forests', 'deep neural networks'])\n", "('NONE', 'XXXXX helps identify patients’ XXXXX and issues', ['distributed cognition', 'interaction strategies'])\n", "('method used for task', 'we proposed a XXXXX to automatically de-identify XXXXX', ['hybrid system', 'electronic medical records'])\n", "('method used for task', 'supervised XXXXX to identify XXXXX for heart disease in ehrs', ['information extraction', 'risk factors'])\n", "('method used for task', 'supervised XXXXX to identify risk factors for XXXXX in ehrs', ['information extraction', 'heart disease'])\n", "('method used for task', 'supervised information extraction to identify XXXXX for XXXXX in ehrs', ['risk factors', 'heart disease'])\n", "('method used for task', 'XXXXX and glass box evaluations are both needed to develop XXXXX', ['black box', 'complex systems'])\n", "('method used for task', 'XXXXX and XXXXX evaluations are both needed to develop complex systems', ['black box', 'glass box'])\n", "('method used for task', 'black box and XXXXX evaluations are both needed to develop XXXXX', ['complex systems', 'glass box'])\n", "('NONE', 'XXXXX of XXXXX using approximate randomization techniques', ['statistical testing', 'model performance'])\n", "('method used for task', 'we used XXXXX ( nlp ) to extract XXXXX', ['natural language processing', 'heart disease risk factors'])\n", "('method used for task', 'developed a rule-based text mining system to identify framingham XXXXX required for predicting XXXXX', ['risk factors', 'coronary artery disease'])\n", "('method used for task', 'developed a rule-based XXXXX to identify framingham XXXXX required for predicting coronary artery disease', ['risk factors', 'text mining system'])\n", "('method used for task', 'developed a rule-based XXXXX to identify framingham risk factors required for predicting XXXXX', ['coronary artery disease', 'text mining system'])\n", "('method used for task', 'XXXXX to identify XXXXX in diabetic patients over time', ['nlp system', 'heart disease risk factors'])\n", "('NONE', 'XXXXX in XXXXX ( emr ) was automated', ['electronic medical records', 'risk factor detection'])\n", "('NONE', 'modeling context with XXXXX yielded better XXXXX', ['distributional semantics', 'predictive models'])\n", "('method used for task', 'we proposed a XXXXX to automatically identify XXXXX', ['hybrid system', 'heart disease risk factors'])\n", "('NONE', 'XXXXX is of great importance for the treatment of XXXXX', ['heart disease', 'risk factor detection'])\n", "('NONE', 'the experimental set up used 6 XXXXX with 7 different XXXXX', ['machine learning models', 'feature sets'])\n", "('NONE', 'fall-injury XXXXX systems are static and provide no form of XXXXX', ['user interaction', 'prevention intervention'])\n", "('method used for task', 'cross-fall XXXXX systems face similar challenges to other fall XXXXX', ['prevention intervention', 'prevention systems'])\n", "('method used for task', 'cross-fall XXXXX systems face similar challenges to other fall XXXXX', ['prevention intervention', 'prevention systems'])\n", "('method used for task', 'the XXXXX can be modified to accommodate XXXXX', ['classification system', 'technological development'])\n", "('method used for task', 'a XXXXX and a hybrid type ii XXXXX are applied', ['fuzzy system', 'data mining technique'])\n", "('method used for task', 'two different types of XXXXX were created to generate a XXXXX', ['membership functions', 'hybrid system'])\n", "('method used for task', 'a XXXXX is established to describe the XXXXX', ['mathematical model', 'localization problem'])\n", "('NONE', 'introduction of XXXXX as a bridge between measurement input and analytical XXXXX for manufacturing', ['repair features', 'cad data'])\n", "('method used for task', 'dynamic determination and adjustment of XXXXX and cax repair process chain execution based on XXXXX', ['function blocks', 'repair features'])\n", "('method used for task', 'dynamic determination and adjustment of repair features and cax XXXXX chain execution based on XXXXX', ['function blocks', 'repair process'])\n", "('method used for task', 'dynamic determination and adjustment of XXXXX and cax XXXXX chain execution based on function blocks', ['repair features', 'repair process'])\n", "('NONE', 'XXXXX study of repairing worn-out turbine blades with repair features and XXXXX', ['use case', 'function blocks'])\n", "('NONE', 'XXXXX study of repairing worn-out turbine blades with XXXXX and function blocks', ['use case', 'repair features'])\n", "('method used for task', 'use case study of repairing worn-out turbine blades with XXXXX and XXXXX', ['function blocks', 'repair features'])\n", "('method used for task', 'in this paper , an XXXXX is proposed for solving minimum time manufacturing XXXXX in multi points manufacturing tasks', ['path planning', 'optimization strategy'])\n", "('method used for task', 'according to the start-stop movement in drilling/spot welding task , the XXXXX problem is converted into a XXXXX ( tsp ) and a series of point to point minimum time transfer XXXXX problems', ['traveling salesman problem', 'path planning'])\n", "('method used for task', 'according to the start-stop movement in drilling/spot welding task , the XXXXX is converted into a XXXXX ( tsp ) and a series of point to point minimum time transfer path planning problems', ['traveling salesman problem', 'path planning problem'])\n", "('method used for task', 'according to the start-stop movement in drilling/spot welding task , the XXXXX problem is converted into a traveling salesman problem ( tsp ) and a series of point to point minimum time transfer XXXXX problems', ['path planning', 'path planning problem'])\n", "('NONE', 'the XXXXX are automatically generated from laser-scanned XXXXX', ['point clouds', 'environment models'])\n", "('NONE', 'performed XXXXX by XXXXX and experimental of live industrial problem', ['design optimization', 'finite element analysis'])\n", "('method used for task', 'this paper enables a “step by step” process to identify the most appropriate XXXXX for a company’s pss problem . an example is also introduced to illustrate the use of the proposed scoring criteria and provide a clear picture of how different XXXXX can be utilized at their best in terms of application', ['design methodology', 'design methodologies'])\n", "('method used for task', 'this paper enables a “step by step” process to identify the most appropriate XXXXX for a company’s pss problem . an example is also introduced to illustrate the use of the proposed XXXXX and provide a clear picture of how different design methodologies can be utilized at their best in terms of application', ['design methodology', 'scoring criteria'])\n", "('NONE', 'this paper enables a “step by step” process to identify the most appropriate design methodology for a company’s pss problem . an example is also introduced to illustrate the use of the proposed XXXXX and provide a clear picture of how different XXXXX can be utilized at their best in terms of application', ['design methodologies', 'scoring criteria'])\n", "('NONE', 'an example is also introduced to illustrate the use of the proposed XXXXX and provide a clear picture of how different XXXXX can be utilized at their best in terms of application', ['design methodologies', 'scoring criteria'])\n", "('method used for task', 'XXXXX and XXXXX scheduling are major issues in optimization of hole-making operations', ['tool travel', 'tool switch'])\n", "('method used for task', 'it is necessary to achieve a correct sequence of hole-making operations to minimize XXXXX and XXXXX and finally to minimize total processing cost', ['tool travel', 'tool switch'])\n", "('NONE', \"we present an efficient technique for sketch-based XXXXX using automatically extracted XXXXX . an automatic and real-time method is proposed to align a user 's hand-drawn sketch line to the contour lines of an image , facilitating a considerable level of ease for XXXXX\", ['3d modeling', 'image features'])\n", "('method used for task', \"we present an efficient technique for sketch-based XXXXX using automatically extracted image features . an automatic and real-time method is proposed to align a user 's hand-drawn sketch line to the XXXXX of an image , facilitating a considerable level of ease for XXXXX\", ['3d modeling', 'contour lines'])\n", "('method used for task', \"we present an XXXXX for sketch-based XXXXX using automatically extracted image features . an automatic and real-time method is proposed to align a user 's hand-drawn sketch line to the contour lines of an image , facilitating a considerable level of ease for XXXXX\", ['3d modeling', 'efficient technique'])\n", "('method used for task', \"we present an efficient technique for sketch-based XXXXX using automatically extracted image features . an automatic and real-time method is proposed to align a user 's hand-drawn XXXXX to the contour lines of an image , facilitating a considerable level of ease for XXXXX\", ['3d modeling', 'sketch line'])\n", "('method used for task', \"we present an efficient technique for sketch-based 3d modeling using automatically extracted XXXXX . an automatic and real-time method is proposed to align a user 's hand-drawn sketch line to the XXXXX of an image , facilitating a considerable level of ease for 3d modeling\", ['image features', 'contour lines'])\n", "('NONE', \"we present an efficient technique for sketch-based XXXXX using automatically extracted XXXXX . an automatic and real-time method is proposed to align a user 's hand-drawn sketch line to the contour lines of an image , facilitating a considerable level of ease for XXXXX\", ['image features', '3d modeling'])\n", "('method used for task', \"we present an XXXXX for sketch-based 3d modeling using automatically extracted XXXXX . an automatic and real-time method is proposed to align a user 's hand-drawn sketch line to the contour lines of an image , facilitating a considerable level of ease for 3d modeling\", ['image features', 'efficient technique'])\n", "('method used for task', \"we present an efficient technique for sketch-based 3d modeling using automatically extracted XXXXX . an automatic and real-time method is proposed to align a user 's hand-drawn XXXXX to the contour lines of an image , facilitating a considerable level of ease for 3d modeling\", ['image features', 'sketch line'])\n", "('method used for task', \"we present an efficient technique for sketch-based XXXXX using automatically extracted image features . an automatic and real-time method is proposed to align a user 's hand-drawn sketch line to the XXXXX of an image , facilitating a considerable level of ease for XXXXX\", ['contour lines', '3d modeling'])\n", "('method used for task', \"we present an XXXXX for sketch-based 3d modeling using automatically extracted image features . an automatic and real-time method is proposed to align a user 's hand-drawn sketch line to the XXXXX of an image , facilitating a considerable level of ease for 3d modeling\", ['contour lines', 'efficient technique'])\n", "('method used for task', \"we present an efficient technique for sketch-based 3d modeling using automatically extracted image features . an automatic and real-time method is proposed to align a user 's hand-drawn XXXXX to the XXXXX of an image , facilitating a considerable level of ease for 3d modeling\", ['contour lines', 'sketch line'])\n", "('method used for task', \"we present an XXXXX for sketch-based XXXXX using automatically extracted image features . an automatic and real-time method is proposed to align a user 's hand-drawn sketch line to the contour lines of an image , facilitating a considerable level of ease for XXXXX\", ['3d modeling', 'efficient technique'])\n", "('method used for task', \"we present an efficient technique for sketch-based XXXXX using automatically extracted image features . an automatic and real-time method is proposed to align a user 's hand-drawn XXXXX to the contour lines of an image , facilitating a considerable level of ease for XXXXX\", ['3d modeling', 'sketch line'])\n", "('method used for task', \"we present an XXXXX for sketch-based 3d modeling using automatically extracted image features . an automatic and real-time method is proposed to align a user 's hand-drawn XXXXX to the contour lines of an image , facilitating a considerable level of ease for 3d modeling\", ['efficient technique', 'sketch line'])\n", "('NONE', 'we use a XXXXX to align a sketch line to the outlines of an image using the features of the sketch line and XXXXX of an image , and some operations are proposed to refine the result of alignment', ['geometric method', 'contour lines'])\n", "('method used for task', 'we use a XXXXX to align a XXXXX to the outlines of an image using the features of the XXXXX and contour lines of an image , and some operations are proposed to refine the result of alignment', ['geometric method', 'sketch line'])\n", "('method used for task', 'we use a XXXXX to align a XXXXX to the outlines of an image using the features of the XXXXX and contour lines of an image , and some operations are proposed to refine the result of alignment', ['geometric method', 'sketch line'])\n", "('NONE', 'we use a geometric method to align a XXXXX to the outlines of an image using the features of the XXXXX and XXXXX of an image , and some operations are proposed to refine the result of alignment', ['contour lines', 'sketch line'])\n", "('NONE', 'we use a geometric method to align a XXXXX to the outlines of an image using the features of the XXXXX and XXXXX of an image , and some operations are proposed to refine the result of alignment', ['contour lines', 'sketch line'])\n", "('NONE', 'in the sketch-based XXXXX , the XXXXX is represented by a editable spline , therefore , the aligned XXXXX can be further adjusted interactively', ['3d modeling method', 'sketch line'])\n", "('NONE', 'in the sketch-based XXXXX , the XXXXX is represented by a editable spline , therefore , the aligned XXXXX can be further adjusted interactively', ['3d modeling method', 'sketch line'])\n", "('NONE', 'XXXXX of the XXXXX on the surface of a sphere', ['numerical solution', 'monge–ampère equation'])\n", "('method used for task', 'the XXXXX applies simultaneously to fluid mechanics and XXXXX', ['mathematical model', 'solid mechanics'])\n", "('method used for task', 'we present an automated ensemble XXXXX for modelling cerebrovascular XXXXX under a range of exercise intensities', ['blood flow', 'simulation method'])\n", "('method used for task', 'software for modeling uncertainty using XXXXX and XXXXX expansions', ['monte carlo', 'polynomial chaos'])\n", "('NONE', 'increasing XXXXX will enhance XXXXX', ['confinement ratio', 'droplet deformation'])\n", "('NONE', 'the droplet is found to orient more towards the flow direction with increasing XXXXX or XXXXX', ['viscosity ratio', 'confinement ratio'])\n", "('method used for task', 'the transparency and robustness are considered as an XXXXX and solved by applying XXXXX', ['optimization problem', 'genetic algorithms'])\n", "('method used for task', 'a XXXXX for 3-d XXXXX is presented', ['mathematical model', 'computer graphs'])\n", "('method used for task', 'a theoretical approach of a XXXXX for 3-d XXXXX and corresponding methods for empirical testing are presented', ['mathematical model', 'computer graphs'])\n", "('method used for task', 'proposing an approach that can provide admissions control for all vod systems . it addresses the following challenges : XXXXX and several algorithms for XXXXX', ['resource allocation', 'scheduling policies'])\n", "('NONE', 'autogrow is an XXXXX that facilitates XXXXX and optimization', ['evolutionary algorithm', 'drug design'])\n", "('NONE', 'the XXXXX leverages strengths of two XXXXX , rtc and railsys', ['hybrid approach', 'simulation tools'])\n", "('method used for task', 'XXXXX to assess the frequency and size of XXXXX', ['empirical analysis', 'coherent clusters'])\n", "('method used for task', 'a study on the relationship between XXXXX and XXXXX', ['coherent clusters', 'system evolution'])\n", "('NONE', 'the approach was assessed in four XXXXX in two parallel XXXXX', ['case studies', 'research projects'])\n", "('method used for task', 'evaluation of the approach includes XXXXX and XXXXX', ['comparative analysis', 'usability studies'])\n", "('method used for task', 'XXXXX is knowledge-intensive and XXXXX is crucial', ['software engineering', 'intellectual capital'])\n", "('method used for task', 'the XXXXX is increased while the XXXXX is reduced', ['classification rate', 'texture feature space'])\n", "('NONE', 'XXXXX based on assessment of XXXXX through domain expert interviews', ['system design', 'user needs'])\n", "('NONE', 'XXXXX via smartphone facilitate the XXXXX', ['collaborative work', 'emergency management'])\n", "('method used for task', 'XXXXX for XXXXX and visual reasoning using taxonomy', ['conceptual graphs', 'knowledge representation'])\n", "('NONE', 'XXXXX improves the XXXXX of the credit ratings', ['classification accuracy', 'feature selection procedure'])\n", "('NONE', 'a XXXXX can enhance the XXXXX between particles', ['crossover operator', 'information exchange'])\n", "('method used for task', 'an efficient XXXXX based without center set according to XXXXX of the network', ['structural properties', 'community detection algorithm'])\n", "('method used for task', 'a simple XXXXX is proposed to construct a XXXXX', ['two-stage approach', 'fuzzy regression model'])\n", "('NONE', 'the XXXXX contains the XXXXX and a fuzzy adjustment term', ['fuzzy model', 'crisp coefficients'])\n", "('method used for task', 'we propose a hybrid artificial bee XXXXX for the cyclic XXXXX', ['colony algorithm', 'antibandwidth problem'])\n", "('NONE', 'the XXXXX can get a XXXXX for small regularization values', ['decomposition method', 'fast convergence'])\n", "('method used for task', 'constructing a XXXXX and XXXXX with heterogeneous items firstly', ['joint replenishment', 'delivery model'])\n", "('NONE', 'a novel XXXXX metric called XXXXX ( gig ) is proposed', ['feature selection', 'global information gain'])\n", "('NONE', 'an XXXXX called maximizing XXXXX ( mgig ) is developed', ['efficient algorithm', 'global information gain'])\n", "('NONE', 'a two-phase cost-sensitive XXXXX combining with XXXXX', ['emotion classification', 'topic detection'])\n", "('method used for task', 'we develop a XXXXX agr-sce to find the XXXXX', ['heuristic algorithm', 'generalization reduct'])\n", "('NONE', 'the XXXXX was utilized to find similar users in the XXXXX', ['social network', 'collaborative filtering'])\n", "('NONE', 'we first analyze the shortages of the existing XXXXX in XXXXX', ['similarity measures', 'collaborative filtering'])\n", "('NONE', 'we demonstrated the occurrence of XXXXX in different XXXXX in dea', ['rank reversal', 'ranking models'])\n", "('NONE', 'our method outperforms state-of-the-art solutions on XXXXX , achieving in particular a XXXXX', ['high precision', 'benchmark data'])\n", "('method used for task', 'it uses a XXXXX to provide personalized treatments to patients with XXXXX and any advices to prevent it', ['recommender system', 'low back pain'])\n", "('method used for task', 'XXXXX is regarded as a multi-index XXXXX', ['feature selection', 'evaluation process'])\n", "('method used for task', 'we apply a two-stage XXXXX to evaluate XXXXX and revenue efficiency', ['cost efficiency', 'dea model'])\n", "('method used for task', 'we develop a XXXXX to deal with uncertain or vague XXXXX', ['fuzzy ontology', 'knowledge representation'])\n", "('NONE', 'the integration of the dual XXXXX gifts the cso algorithm with powerful XXXXX', ['search mechanisms', 'global search ability'])\n", "('method used for task', 'new XXXXX models are based on full XXXXX', ['probability density functions', 'key profile'])\n", "('NONE', 'we construct a margin related XXXXX to learn the weights of XXXXX', ['loss function', 'base classifiers'])\n", "('NONE', 'the system was improved with XXXXX , XXXXX and statistical techniques', ['text mining', 'neural networks'])\n", "('method used for task', 'the system was improved with XXXXX , neural networks and XXXXX', ['text mining', 'statistical techniques'])\n", "('method used for task', 'the system was improved with text mining , XXXXX and XXXXX', ['neural networks', 'statistical techniques'])\n", "('NONE', 'new design for reliable XXXXX with higher XXXXX and quality', ['inference system', 'software reliability'])\n", "('method used for task', 'introduces a novel XXXXX based on bagged and XXXXX', ['fuzzy clustering', 'segmentation method'])\n", "('method used for task', 'proposes how to adapt bagged XXXXX for XXXXX', ['fuzzy clustering', 'fuzzy data'])\n", "('NONE', 'a fusion of iso-clahe improved the XXXXX from a low lighting or XXXXX', ['contrast ratio', 'eye image'])\n", "('method used for task', 'an innovative XXXXX is proposed for diagnosing XXXXX', ['intelligent system', 'heart diseases'])\n", "('method used for task', 'ctc combined with smote tops state of the XXXXX designed to tackle XXXXX', ['class imbalance', 'art techniques'])\n", "('NONE', 'an XXXXX of XXXXX and classification methods is developed', ['experimental study', 'feature extraction'])\n", "('NONE', 'an XXXXX of feature extraction and XXXXX is developed', ['experimental study', 'classification methods'])\n", "('method used for task', 'an experimental study of XXXXX and XXXXX is developed', ['feature extraction', 'classification methods'])\n", "('method used for task', 'XXXXX conclude that isso outperforms abc for 50 XXXXX', ['numerical examples', 'benchmark functions'])\n", "('NONE', 'novel sudden XXXXX index ( scdi ) is proposed using XXXXX', ['ecg signals', 'cardiac death'])\n", "('method used for task', 'a multi-attribute XXXXX is proposed by combining XXXXX and topsis', ['relative entropy', 'ranking method'])\n", "('method used for task', 'XXXXX and XXXXX based mil algorithm is proposed', ['extreme learning machine', 'classifier ensemble'])\n", "('NONE', 'a new method of XXXXX of XXXXX is presented', ['automatic verification', 'knowledge base'])\n", "('method used for task', 'verification looks for discrepancy between the XXXXX and XXXXX', ['expert knowledge', 'knowledge base'])\n", "('NONE', 'experts and XXXXX system use the half-marks in XXXXX', ['automatic verification', 'inference rules'])\n", "('method used for task', 'we propose a fast XXXXX to solve the modeled XXXXX', ['memetic algorithm', 'optimization problems'])\n", "('method used for task', 'a XXXXX is used to show the applicability of the proposed XXXXX', ['case study', 'optimization model'])\n", "('method used for task', 'XXXXX as regards key parameters is performed in the XXXXX', ['sensitivity analysis', 'case study'])\n", "('method used for task', 'the approach is based on XXXXX and XXXXX', ['semantic technologies', 'web standards'])\n", "('NONE', 'a XXXXX describes experiences with a XXXXX in industrial use for years', ['case study', 'decision support system'])\n", "('NONE', 'XXXXX and classification of XXXXX using hadoop framework', ['feature selection', 'microarray data'])\n", "('method used for task', 'mapreduce based XXXXX are proposed for XXXXX ( fs )', ['statistical tests', 'feature selection'])\n", "('method used for task', 'XXXXX are converted to 1d signals using XXXXX', ['fundus images', 'radon transform'])\n", "('method used for task', 'the novel node coupling XXXXX for XXXXX are proposed', ['clustering methods', 'link prediction'])\n", "('method used for task', 'we exploit XXXXX shapelets for complex XXXXX', ['time series', 'human activity recognition'])\n", "('method used for task', 'we implement a XXXXX based on smartphone for XXXXX', ['human activity recognition', 'prototype system'])\n", "('NONE', 'ldmdba has a XXXXX of ( logdlogn ) for predicting a new XXXXX', ['time complexity', 'data point'])\n", "('NONE', 'we apply XXXXX in XXXXX to make the best use of historical driving data', ['data mining algorithms', 'train operations'])\n", "('NONE', 'two sto algorithms are proposed by combining XXXXX , XXXXX and train parking methods', ['expert knowledge', 'data mining'])\n", "('method used for task', 'XXXXX and XXXXX are conducted to present the findings', ['numerical simulation', 'empirical studies'])\n", "('NONE', 'a new total XXXXX in XXXXX is proposed', ['uncertainty measure', 'evidence theory'])\n", "('method used for task', 'the rimer-based XXXXX is fine-tuned and validated using XXXXX', ['prediction model', 'historical data'])\n", "('method used for task', 'a novel XXXXX is used to enhance the XXXXX', ['learning strategy', 'global search ability'])\n", "('NONE', 'it proposes a XXXXX qos dynamic XXXXX based on improvedâ case-based reasoning ( cbr )', ['web service', 'prediction method'])\n", "('method used for task', 'it proposes a XXXXX qos dynamic prediction method based on improvedâ XXXXX ( cbr )', ['web service', 'case-based reasoning'])\n", "('method used for task', 'it proposes a web service qos dynamic XXXXX based on improvedâ XXXXX ( cbr )', ['prediction method', 'case-based reasoning'])\n", "('method used for task', 'an opposite-based approach to XXXXX is presented as a unifying view to basic XXXXX', ['fuzzy modeling', 'knowledge representation'])\n", "('method used for task', 'we claim that paired concepts are the basic XXXXX for XXXXX', ['building blocks', 'knowledge representation'])\n", "('NONE', 'we extract a XXXXX from event logs using XXXXX', ['process mining', 'process structure'])\n", "('NONE', 'XXXXX improves the efficiency in XXXXX', ['sample selection', 'feature dimension reduction'])\n", "('NONE', 'bayesian XXXXX of a XXXXX using a novel mcmc algorithm', ['system identification', 'nonlinear dynamical system'])\n", "('NONE', 'data annealing is applied to the XXXXX of a XXXXX', ['system identification', 'nonlinear dynamical system'])\n", "('NONE', 'techniques are developed which allow one to choose a subset of the available training data which is ‘highly informative’ with regards to both levels of XXXXX – XXXXX and model selection', ['bayesian inference', 'parameter estimation'])\n", "('method used for task', 'techniques are developed which allow one to choose a subset of the available training data which is ‘highly informative’ with regards to both levels of XXXXX – parameter estimation and XXXXX', ['bayesian inference', 'model selection'])\n", "('method used for task', 'techniques are developed which allow one to choose a subset of the available training data which is ‘highly informative’ with regards to both levels of bayesian inference – XXXXX and XXXXX', ['parameter estimation', 'model selection'])\n", "('NONE', 'virtual and XXXXX of the controllers on a fully XXXXX', ['experimental validation', 'electric vehicle'])\n", "('NONE', 'a XXXXX of myocardial XXXXX imaging is proposed', ['finite element model', 'mr perfusion'])\n", "('method used for task', 'XXXXX are evaluated directly from the 4d XXXXX , without the need of any computational mesh', ['pressure differences', 'flow data'])\n", "('NONE', 'we present a standardisation for evaluating XXXXX for late enhancement imaging of infarct in the XXXXX', ['left ventricle', 'segmentation algorithms'])\n", "('method used for task', 'we provide a consensus ground truth obtained with XXXXX on the XXXXX . future algorithms can thus be benchmarked to provide a more reliable result', ['statistical modelling', 'benchmark datasets'])\n", "('NONE', 'the XXXXX has a positive correlation with the XXXXX', ['tensile strength', 'strain rate'])\n", "('NONE', 'XXXXX in the bulk solder plays dominant role at the low XXXXX', ['ductile fracture', 'strain rate'])\n", "('NONE', 'an XXXXX on bond wire looping by using high-speed XXXXX was presented', ['experimental study', 'video analysis'])\n", "('NONE', 'compare scanning acoustic XXXXX with the XXXXX throughout a power cycling test which help to build correlation between the two methods', ['structure function', 'microscopy images'])\n", "('method used for task', 'XXXXX and XXXXX are applied for improvement', ['arq dynamics', 'crc codes'])\n", "('method used for task', 'XXXXX and XXXXX are applied for improvement', ['arq dynamics', 'crc codes'])\n", "('method used for task', 'an efficient XXXXX for a dendrite morphological XXXXX', ['training algorithm', 'neural network'])\n", "('NONE', 'the proposed algorithm is much less XXXXX than the trace norm minimization algorithm especially facing the XXXXX', ['computational cost', 'large data'])\n", "('NONE', 'we proposed a XXXXX by using a selective XXXXX technique to reduce dsp complexity', ['mimo system', 'mode excitation'])\n", "('method used for task', 'a domain specific language and XXXXX for on- and off-line XXXXX', ['data analytics', 'runtime models'])\n", "('NONE', 'our method maximises the XXXXX between a small set of XXXXX', ['euclidean distance', 'feature points'])\n", "('method used for task', 'ensemble XXXXX based on XXXXX of texture features was effective', ['svm classification', 'sparse coding'])\n", "('NONE', 'ensemble XXXXX based on sparse coding of XXXXX was effective', ['svm classification', 'texture features'])\n", "('NONE', 'ensemble svm classification based on XXXXX of XXXXX was effective', ['sparse coding', 'texture features'])\n", "('NONE', 'we utilise a novel social connections metric and a XXXXX as XXXXX', ['contextual information', 'scene model'])\n", "('method used for task', 'a XXXXX tool for robotized XXXXX has been developed and constructed', ['cable feeder', 'cable winding'])\n", "('method used for task', 'development of a framework for XXXXX and in-service data feedback to XXXXX', ['knowledge management', 'product development'])\n", "('method used for task', 'demonstrating how XXXXX is captured , fed back and reused for XXXXX', ['field data', 'product development'])\n", "('NONE', 'the paper provides XXXXX of path trajectory generation and ndt XXXXX on two large , curved surfaces of a composite aerofoil component', ['experimental validation', 'data acquisition'])\n", "('NONE', 'the paper provides XXXXX of path trajectory generation and ndt data acquisition on two large , XXXXX of a composite aerofoil component', ['experimental validation', 'curved surfaces'])\n", "('NONE', 'the paper provides experimental validation of path trajectory generation and ndt XXXXX on two large , XXXXX of a composite aerofoil component', ['data acquisition', 'curved surfaces'])\n", "('method used for task', 'XXXXX for multi-level XXXXX', ['hierarchical model', 'adaptive systems'])\n", "('NONE', 'application to the XXXXX of atvs XXXXX', ['case study', 'motion control'])\n", "('method used for task', 'XXXXX is robust to XXXXX and video frame-based attacks', ['signal processing', 'watermark scheme'])\n", "('NONE', 'XXXXX of costas loop in the signal׳s XXXXX is demonstrated', ['nonlinear analysis', 'phase space'])\n", "('NONE', 'an adaptive multi-view feature selection ( amfs ) algorithm is proposed to fuse XXXXX formotion XXXXX', ['multiple features', 'data retrieval'])\n", "('method used for task', 'XXXXX for the binary m-qim ( multi-symbol XXXXX modulation )', ['theoretical framework', 'quantization index'])\n", "('method used for task', 'the difficulty in XXXXX is to balance the simulation cost and XXXXX', ['spot market', 'execution time'])\n", "('NONE', 'an analysis of key drivers governing XXXXX on XXXXX has been carried out', ['amazon ec2', 'spot prices'])\n", "('method used for task', 'we give a survey of 24 XXXXX for text-dependent XXXXX', ['speaker verification', 'speech databases'])\n", "('method used for task', 'we present a method for interpolation between XXXXX for XXXXX', ['speech synthesis', 'language varieties'])\n", "('method used for task', 'a layered algorithm integrates XXXXX and XXXXX', ['dynamic bayesian network', 'supervised clustering'])\n", "('method used for task', 'XXXXX shows the great value to XXXXX in data mining', ['supervised clustering', 'feature reduction'])\n", "('method used for task', 'XXXXX shows the great value to feature reduction in XXXXX', ['supervised clustering', 'data mining'])\n", "('NONE', 'supervised clustering shows the great value to XXXXX in XXXXX', ['feature reduction', 'data mining'])\n", "('method used for task', 'XXXXX for transport mode detection studies from XXXXX using coefficient of variation cv', ['sample size', 'gps data'])\n", "('NONE', 'the proposed XXXXX ( ma ) is a combination of a XXXXX and a simulated annealing approach', ['memetic algorithm', 'genetic algorithm'])\n", "('method used for task', 'a XXXXX for evaluating XXXXX locations is proposed', ['generic model', 'traffic accident'])\n" ] } ], "source": [ "kb_entpairs, unlab_sents, unlab_entpairs = ie.readDataForDistantSupervision()\n", "print(len(kb_entpairs), \"'KB' entity pairs for relation 'method used for task' :\", kb_entpairs[0:5])\n", "print(len(unlab_entpairs), 'all entity pairs')\n", "ie.distantlySupervisedExtraction(kb_entpairs, unlab_sents, unlab_entpairs, testing_patterns, testing_entpairs)" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "* The results we get here are the same as for supervised relation extraction. This is because the distant supervision heuristic identified the same positive and negative training examples as in the manually labelled dataset\n", "* In practice, the distant supervision heuristic typically leads to noisy training data due to several reasons\n", " * Overlapping relations\n", " * For instance, 'employee-of' entails 'lecturer-at' and there are some overlapping entity pairs between the relations 'employee-of' and 'student-at'" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "* In practice, the distant supervision heuristic typically leads to noisy training data due to several reasons\n", " * The next problem is ambiguous entities\n", " * e.g. 'EM' has many possible meanings, only one of which is 'Expectation Maximisation', see [the Wikipedia disambiguation page for the acronym](https://en.wikipedia.org/wiki/EM).\n", " * Not every sentence an entity pair that is a positive example for a relation appears in actually contains that relation" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "* In practice, the distant supervision heuristic typically leads to noisy training data due to several reasons\n", " * Ambiguous entities\n", " * e.g. compare the sentence from [the Wikipedia EM definition](https://en.wikipedia.org/wiki/Expectation%E2%80%93maximization_algorithm) \n", " 'Expectation–maximization algorithm, an algorithm for finding maximum likelihood estimates (...)' with \n", " 'In this section we introduce EM (...)'\n", " * The first one is a true mention of 'method used for task', whereas the second one is not." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "## Universal Schema\n", "* Recall that for the pattern-based and bootstrapping approaches earlier, we were looking for simplified paths between entity pairs $\\mathcal{E}$ expressing a certain relation $\\mathcal{R}$ which we defined beforehand\n", " * This **restricts the relation extraction problem to known relation types** $\\mathcal{R}$\n", " * In order to overcome that limitation, we could have defined new relations on the spot and added them to $\\mathcal{R}$ by introducing new relation types for certain simplified paths between entity pairs" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "## Universal Schema\n", "* The goal of universal schemas is to overcome the limitation of having to pre-define relations, but within the supervised learning paradigm\n", "* This is possible by thinking of paths between entity pairs **as relation expressions themselves**\n", "* Simplified paths between entity pairs and relation labels are no longer considered separately, but instead the paths between entity pairs and relations is **modelled in the same space**\n" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "The space of entity pairs and relations is defined by a matrix:\n", "\n", "| | demonstrates XXXXX for XXXXXX | XXXXX is capable of XXXXXX | an XXXXX model is employed for XXXXX | XXXXX decreases the XXXXX | method is used for task |\n", "| ------ | ----------- |\n", "| 'text mining', 'building domain ontology' | 1 | | | | 1 |\n", "| 'ensemble classifier', 'detection of construction materials' | | | 1 | | 1 |\n", "| 'data mining', 'characterization of wireless systems performance'| | 1 | | | ? |\n", "| 'frequency domain', 'computational cost' | | | | 1 | ? |" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "* Here, 'method is used for task' is a relation defined by a KB schema\n", "* the other relations are surface pattern relations generated by blanking entity pairs in sentences\n", "* Where an entity pair and a KB relation or surface pattern relation co-occur, this is signified by a '1'\n", "* For some of the entities and surface pairs, a label for 'method used for task' is available, whereas for others, it is not (signified by the '?')" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "* The task is to turn the '?'s into 0/1 predictions for the 'method for task' relation\n", "* Note that we can use the same data this is the same task of for relation extraction extraction as considered previously with supervised learning, merely the data representation and model are different" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "* In order to solve this prediction task, we will learn to fill in the empty cells in the matrix\n", "* This is achieved by learning to distinguish between entity pairs and relations which co-occur in our training data and entity pairs and relations which are not known to co-occur (the empty cells)" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "* Each training instance consists of a surface pattern or KB relation $\\mathcal{r_{pos}}$ and an entity pair $\\mathcal{e_{pos}}$ the relation co-occurs with, as well as a relation $\\mathcal{r_{neg}}$ and a entity pair $\\mathcal{e_{neg}}$ that do not co-occur in the training data\n", "* The positive relations and entity pairs are directly taken from the annotated data\n", "* The negative entity pairs and relations are sampled randomly from data points which are represented by the empty cell in the matrix above. The goal is to estimate, for a relation $\\mathcal{r}$ such as 'method is used for task' and an unseen entity pair such as $\\mathcal{e}$, e.g. ('frequency domain', 'computational cost'), what the probability $\\mathcal{p(y_{r,e} = 1)}$ is." ] }, { "cell_type": "code", "execution_count": 24, "metadata": { "slideshow": { "slide_type": "subslide" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Max relation length: 16\n", "Max entity pair length: 9\n" ] } ], "source": [ "# data reading\n", "training_sents, training_entpairs, training_labels = ie.readLabelledData()\n", "\n", "# split positive and negative training data\n", "pos_train_ids, neg_train_ids = ie.split_labels_pos_neg(training_labels + training_labels)\n", "\n", "training_toks_pos = [t.split(\" \") for i, t in enumerate(training_sents + training_labels) if i in pos_train_ids]\n", "training_toks_neg = [t.split(\" \") for i, t in enumerate(training_sents + training_labels) if i in neg_train_ids]\n", "\n", "training_ent_toks_pos = [\" || \".join(t).split(\" \") for i, t in enumerate(training_entpairs + training_entpairs) if i in pos_train_ids]\n", "training_ent_toks_neg = [\" || \".join(t).split(\" \") for i, t in enumerate(training_entpairs + training_entpairs) if i in neg_train_ids]\n", "testing_ent_toks = [\" || \".join(t).split(\" \") for t in testing_entpairs]\n", "\n", "# print length statistics\n", "lens_rel = [len(s) for s in training_toks_pos + training_toks_neg]\n", "lens_ents = [len(s) for s in training_ent_toks_pos + training_ent_toks_neg + testing_ent_toks]\n", "print(\"Max relation length:\", max(lens_rel))\n", "print(\"Max entity pair length:\", max(lens_ents))" ] }, { "cell_type": "code", "execution_count": 25, "metadata": { "slideshow": { "slide_type": "subslide" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Final vocab size: 163\n", "Final vocab size: 138\n", "{0: 'UNK', 1: 'XXXXX', 2: 'for', 3: 'used', 4: 'method', 5: 'task', 6: 'NONE', 7: 'the', 8: 'a', 9: 'of', 10: 'and', 11: 'to', 12: 'is', 13: 'in', 14: 'are', 15: 'model', 16: 'proposed', 17: 'we', 18: 'this', 19: 'new', 20: 'presented', 21: 'as', 22: 'with', 23: 'propose', 24: 'pso-based', 25: 'anfis', 26: 'approaches', 27: 'affective', 28: 'combination', 29: '(', 30: ')', 31: 'on', 32: 'using', 33: 'demonstrates', 34: 'paper', 35: 'proposes', 36: 'solve', 37: 'design', 38: 'an', 39: 'swarm', 40: 'intelligence', 41: 'introduced', 42: 'clustering', 43: 'techniques', 44: 'text', 45: 'mining', 46: 'building', 47: 'able', 48: 'enhance', 49: 'fully', 50: '3d', 51: 'buildings', 52: 'two', 53: 'more', 54: 'capable', 55: 'product', 56: 'customer', 57: 'satisfaction', 58: 'solved', 59: 'obtained', 60: 'optimal', 61: 'section', 62: 'shape', 63: 'sizing', 64: 'cable–truss', 65: 'structures', 66: 'employed', 67: 'assists', 68: 'chaos', 69: 'theory', 70: 'here', 71: 'use', 72: 'partially', 73: 'converged', 74: 'data', 75: 'construct', 76: 'ga', 77: 'automated', 78: 'two-stage', 79: 'speed', 80: 'reducer', 81: 'between', 82: 'parameters', 83: 'response', 84: 'adopted', 85: 'vns', 86: 'application', 87: 'grid', 88: 'portioning', 89: 'associated', 90: 'damage', 91: 'assessment', 92: 'sa', 93: 'helps', 94: 'ivs', 95: 'time', 96: 'constraint', 97: 'study', 98: 'focused', 99: 'images', 100: 'research', 101: 'aims', 102: 'reduce', 103: 'overall', 104: 'collision', 105: 'checks', 106: 'virtual', 107: 'unified', 108: 'methodology', 109: 'operations', 110: 'generate', 111: 'that', 112: 'networks', 113: 'all', 114: 'except', 115: 'iteration', 116: 'times', 117: 'self-adapted', 118: 'need', 119: 'not', 120: 'be', 121: 'tuned', 122: 'leverages', 123: 'empirical', 124: 'data-mining', 125: 'algorithms', 126: 'rough', 127: 'set', 128: 'innovatively', 129: 'establishing', 130: 'judging', 131: 'metrics', 132: 'estimate', 133: 'expected', 134: 'laser', 135: 'scan', 136: 'can', 137: 'ensure', 138: 'dynamic', 139: 'construction', 140: 'sites', 141: 'travelling', 142: 'completed', 143: 'by', 144: 'fds', 145: 'practical', 146: 'approach', 147: 'applications', 148: 'software', 149: 'listed', 150: 'ann', 151: 'find', 152: 'welding', 153: 'obtaining', 154: 'extended', 155: 'our', 156: 'identification', 157: 'smoothing', 158: 'decreases', 159: 'specified', 160: 'huge', 161: 'impact', 162: 'allowing'}\n" ] } ], "source": [ "# vectorise data (assign IDs to words)\n", "count_rels, dictionary_rels, reverse_dictionary_rels = ie.build_dataset(\n", " [token for senttoks in training_toks_pos + training_toks_neg for token in senttoks])\n", "\n", "count_ents, dictionary_ents, reverse_dictionary_ents = ie.build_dataset(\n", " [token for senttoks in training_ent_toks_pos + training_ent_toks_neg for token in senttoks])\n", "\n", "print(reverse_dictionary_rels)" ] }, { "cell_type": "code", "execution_count": 26, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "[33 1 10 42 43 2 1 0 0 0 0 0 0 0 0 0] \n", " [33 44 45 10 1 2 46 1 0 0 0 0 0 0 0 0]\n" ] } ], "source": [ "# transform sentences to IDs, pad vectors for each sentence so they have same length\n", "rels_train_pos = [ie.transform_dict(dictionary_rels, senttoks, max(lens_rel)) for senttoks in training_toks_pos]\n", "rels_train_neg = [ie.transform_dict(dictionary_rels, senttoks, max(lens_rel)) for senttoks in training_toks_neg]\n", "ents_train_pos = [ie.transform_dict(dictionary_ents, senttoks, max(lens_ents)) for senttoks in training_ent_toks_pos]\n", "ents_train_neg = [ie.transform_dict(dictionary_ents, senttoks, max(lens_ents)) for senttoks in training_ent_toks_neg]\n", "\n", "print(rels_train_pos[0], \"\\n\", rels_train_pos[1])" ] }, { "cell_type": "code", "execution_count": 27, "metadata": { "slideshow": { "slide_type": "subslide" } }, "outputs": [], "source": [ "# Negatively sample some entity pairs for training. Here we have some manually labelled neg ones, so we can sample from them.\n", "ents_train_neg_samp = [random.choice(ents_train_neg) for _ in rels_train_neg]\n", " \n", "ents_test_pos = [ie.transform_dict(dictionary_ents, senttoks, max(lens_ents)) for senttoks in testing_ent_toks]\n", "# Sample those test entity pairs from the training ones as for those we have neg annotations\n", "ents_test_neg_samp = [random.choice(ents_train_neg) for _ in ents_test_pos] \n", "\n", "vocab_size_rels = len(dictionary_rels)\n", "vocab_size_ents = len(dictionary_ents) \n", "\n", "# for testing, we want to check if each unlabelled instance expresses the given relation \"method for task\"\n", "rels_test_pos = [ie.transform_dict(dictionary_rels, training_toks_pos[-1], max(lens_rel)) for _ in testing_patterns]\n", "rels_test_neg_samp = [random.choice(rels_train_neg) for _ in rels_test_pos]" ] }, { "cell_type": "code", "execution_count": 28, "metadata": { "slideshow": { "slide_type": "subslide" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Max relation length: 16\n", "Max entity pair length: 9\n", "Final vocab size: 163\n", "Final vocab size: 138\n" ] } ], "source": [ "data = ie.vectorise_data(training_sents, training_entpairs, training_labels, testing_patterns, testing_entpairs)\n", "\n", "rels_train_pos, rels_train_neg, ents_train_pos, ents_train_neg_samp, rels_test_pos, rels_test_neg_samp, \\\n", " ents_test_pos, ents_test_neg_samp, vocab_size_rels, vocab_size_ents, max_lens_rel, max_lens_ents, \\\n", " dictionary_rels_rev, dictionary_ents_rev = data\n", " \n", "# setting hyper-parameters\n", "batch_size = 4\n", "repr_dim = 30 # dimensionality of relation and entity pair vectors\n", "learning_rate = 0.001\n", "max_epochs = 31" ] }, { "cell_type": "code", "execution_count": 29, "metadata": { "slideshow": { "slide_type": "subslide" } }, "outputs": [], "source": [ "def create_model_f_reader(max_rel_seq_length, max_cand_seq_length, repr_dim, vocab_size_rels, vocab_size_cands):\n", " \"\"\"\n", " Create a Model F Universal Schema reader (Tensorflow graph).\n", " Args:\n", " max_rel_seq_length: maximum sentence sequence length\n", " max_cand_seq_length: maximum candidate sequence length\n", " repr_dim: dimensionality of vectors\n", " vocab_size_rels: size of relation vocabulary\n", " vocab_size_cands: size of candidate vocabulary\n", " Returns:\n", " dotprod_pos: dot product between positive entity pairs and relations\n", " dotprod_neg: dot product between negative entity pairs and relations\n", " diff_dotprod: difference in dot product of positive and negative instances, used for BPR loss (optional)\n", " [relations_pos, relations_neg, ents_pos, ents_neg]: placeholders, fed in during training for each batch\n", " \"\"\"" ] }, { "cell_type": "code", "execution_count": 30, "metadata": { "slideshow": { "slide_type": "subslide" } }, "outputs": [], "source": [ "# Placeholders (empty Tensorflow variables) for positive and negative relations and entity pairs\n", "# In each training epoch, for each batch, those will be set through mini batching\n", "\n", "relations_pos = tf.placeholder(tf.int32, [None, max_lens_rel], name='relations_pos') # [batch_size, max_rel_seq_len]\n", "relations_neg = tf.placeholder(tf.int32, [None, max_lens_rel], name='relations_neg') # [batch_size, max_rel_seq_len]\n", "\n", "ents_pos = tf.placeholder(tf.int32, [None, max_lens_ents], name=\"ents_pos\") # [batch_size, max_ent_seq_len]\n", "ents_neg = tf.placeholder(tf.int32, [None, max_lens_ents], name=\"ents_neg\") # [batch_size, max_ent_seq_len]" ] }, { "cell_type": "code", "execution_count": 31, "metadata": { "slideshow": { "slide_type": "subslide" } }, "outputs": [], "source": [ "# Creating latent representations of relations and entity pairs\n", "# latent feature representation of all relations, which are initialised randomly\n", "relation_embeddings = tf.Variable(tf.random_uniform([vocab_size_rels, repr_dim], -0.1, 0.1, dtype=tf.float32),\n", " name='rel_emb', trainable=True)\n", "\n", "# latent feature representation of all entity pairs, which are initialised randomly\n", "ent_embeddings = tf.Variable(tf.random_uniform([vocab_size_ents, repr_dim], -0.1, 0.1, dtype=tf.float32),\n", " name='cand_emb', trainable=True)\n", "\n", "# look up latent feature representation for relations and entities in current batch\n", "rel_encodings_pos = tf.nn.embedding_lookup(relation_embeddings, relations_pos)\n", "rel_encodings_neg = tf.nn.embedding_lookup(relation_embeddings, relations_neg)\n", "\n", "ent_encodings_pos = tf.nn.embedding_lookup(ent_embeddings, ents_pos)\n", "ent_encodings_neg = tf.nn.embedding_lookup(ent_embeddings, ents_neg)" ] }, { "cell_type": "code", "execution_count": 32, "metadata": { "slideshow": { "slide_type": "subslide" } }, "outputs": [], "source": [ "# our feature representation here is a vector for each word in a relation or entity \n", "# because our training data is so small\n", "# we therefore take the sum of those vectors to get a representation of each relation or entity pair\n", "rel_encodings_pos = tf.reduce_sum(rel_encodings_pos, 1) # [batch_size, num_rel_toks, repr_dim]\n", "rel_encodings_neg = tf.reduce_sum(rel_encodings_neg, 1) # [batch_size, num_rel_toks, repr_dim]\n", "\n", "ent_encodings_pos = tf.reduce_sum(ent_encodings_pos, 1) # [batch_size, num_ent_toks, repr_dim]\n", "ent_encodings_neg = tf.reduce_sum(ent_encodings_neg, 1) # [batch_size, num_ent_toks, repr_dim]" ] }, { "cell_type": "code", "execution_count": 33, "metadata": { "slideshow": { "slide_type": "subslide" } }, "outputs": [], "source": [ "# measuring compatibility between positive entity pairs and relations\n", "# used for ranking test data\n", "dotprod_pos = tf.reduce_sum(tf.multiply(ent_encodings_pos, rel_encodings_pos), 1)\n", "\n", "# measuring compatibility between negative entity pairs and relations\n", "dotprod_neg = tf.reduce_sum(tf.multiply(ent_encodings_neg, rel_encodings_neg), 1)\n", "\n", "# difference in dot product of positive and negative instances\n", "# used for BPR loss (ranking loss)\n", "diff_dotprod = tf.reduce_sum(tf.multiply(ent_encodings_pos, rel_encodings_pos) - tf.multiply(ent_encodings_neg, rel_encodings_neg), 1)\n" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "To train this model, we define a loss, which tries to maximise the distance between the positive and negative instances. One possibility of this is the logistic loss.\n", "\n", "$\\mathcal{\\sum - log(v_{e_{pos}} * a_{r_{pos}})} + {\\sum log(v_{e_{neg}} * a_{r_{neg}}))}$" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "Now that we have read in the data, vectorised it and created the universal schema relation extraction model, let's start training" ] }, { "cell_type": "code", "execution_count": 34, "metadata": { "slideshow": { "slide_type": "subslide" } }, "outputs": [], "source": [ "# create a the model / Tensorflow computation graph\n", "dotprod_pos, dotprod_neg, diff_dotprod, placeholders = ie.create_model_f_reader(max_lens_rel, max_lens_ents, repr_dim, vocab_size_rels,\n", " vocab_size_ents)\n", "\n", "# logistic loss\n", "loss = tf.reduce_sum(tf.nn.softplus(-dotprod_pos)+tf.nn.softplus(dotprod_neg))\n", "\n", "# alternative: BPR loss\n", "#loss = tf.reduce_sum(tf.nn.softplus(diff_dotprod))" ] }, { "cell_type": "code", "execution_count": 35, "metadata": { "slideshow": { "slide_type": "subslide" } }, "outputs": [], "source": [ "data = [np.asarray(rels_train_pos), np.asarray(rels_train_neg), np.asarray(ents_train_pos), np.asarray(ents_train_neg_samp)]\n", "data_test = [np.asarray(rels_test_pos), np.asarray(rels_test_neg_samp), np.asarray(ents_test_pos), np.asarray(ents_test_neg_samp)]\n", "\n", "# define an optimiser. Here, we use the Adam optimiser\n", "optimizer = tf.train.AdamOptimizer(learning_rate)\n", " \n", "# training with mini-batches\n", "batcher = tfutil.BatchBucketSampler(data, batch_size)\n", "batcher_test = tfutil.BatchBucketSampler(data_test, 1, test=True)" ] }, { "cell_type": "code", "execution_count": 36, "metadata": { "slideshow": { "slide_type": "subslide" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Epoch 1 \tLoss 46.5499911308\n", "Epoch 2 \tLoss 32.969190836\n", "Epoch 3 \tLoss 24.4356733561\n", "Epoch 4 \tLoss 16.987249732\n", "Epoch 5 \tLoss 12.225643754\n", "Epoch 6 \tLoss 8.93038511276\n", "Epoch 7 \tLoss 6.68729266524\n", "Epoch 8 \tLoss 4.87118721008\n", "Epoch 9 \tLoss 3.77563881874\n", "Epoch 10 \tLoss 3.15554860234\n", "Epoch 11 \tLoss 2.62767970562\n", "Epoch 12 \tLoss 2.10903429985\n", "Epoch 13 \tLoss 1.7483137697\n", "Epoch 14 \tLoss 1.47833425552\n", "Epoch 15 \tLoss 1.22717268765\n", "Epoch 16 \tLoss 1.12230950594\n", "Epoch 17 \tLoss 1.04518919438\n", "Epoch 18 \tLoss 0.888099145144\n", "Epoch 19 \tLoss 0.767774961889\n", "Epoch 20 \tLoss 0.717018354684\n", "Epoch 21 \tLoss 0.673434060067\n", "Epoch 22 \tLoss 0.581763647497\n", "Epoch 23 \tLoss 0.545647699386\n", "Epoch 24 \tLoss 0.502182915807\n", "Epoch 25 \tLoss 0.463427852839\n", "Epoch 26 \tLoss 0.414758887142\n", "Epoch 27 \tLoss 0.403060832992\n", "Epoch 28 \tLoss 0.376163519919\n", "Epoch 29 \tLoss 0.336928585544\n", "Epoch 30 \tLoss 0.309169012122\n" ] } ], "source": [ "with tf.Session() as sess:\n", " trainer = tfutil.Trainer(optimizer, max_epochs)\n", " trainer(batcher=batcher, placeholders=placeholders, loss=loss, session=sess)\n", "\n", " # we obtain test scores\n", " test_scores = trainer.test(batcher=batcher_test, placeholders=placeholders, model=tf.nn.sigmoid(dotprod_pos), session=sess)" ] }, { "cell_type": "code", "execution_count": 37, "metadata": { "slideshow": { "slide_type": "subslide" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "\n", "Test predictions by decreasing probability:\n", "0.999502\tUNK optimization problem || UNK optimization problem\tREL\tmethod used for task\n", "0.999185\tgenetic algorithm || optimization problem\tREL\tmethod used for task\n", "0.999185\toptimization problem || genetic algorithm\tREL\tmethod used for task\n", "0.999185\toptimization problem || genetic algorithm\tREL\tmethod used for task\n", "0.998879\tUNK search algorithm || hybrid system\tREL\tmethod used for task\n", "0.998851\tUNK optimization problem || optimal UNK UNK problem\tREL\tmethod used for task\n", "0.998722\tUNK set || topology optimization\tREL\tmethod used for task\n", "0.998654\thybrid method || dynamic system\tREL\tmethod used for task\n", "0.998432\tgenetic algorithm || UNK swarm optimization\tREL\tmethod used for task\n", "0.998432\tgenetic algorithm || UNK swarm optimization\tREL\tmethod used for task\n", "0.998432\tgenetic algorithm || UNK swarm optimization\tREL\tmethod used for task\n", "0.998432\tgenetic algorithm || UNK swarm optimization\tREL\tmethod used for task\n", "0.998426\thybrid method || UNK set theory\tREL\tmethod used for task\n", "0.998383\thybrid method || the neural networks\tREL\tmethod used for task\n", "0.998084\tgenetic algorithm || search algorithm\tREL\tmethod used for task\n", "0.998084\tUNK search algorithm || genetic algorithm\tREL\tmethod used for task\n", "0.998017\tUNK search || hybrid genetic algorithm\tREL\tmethod used for task\n", "0.998007\tgenetic UNK system || genetic\tREL\tmethod used for task\n", "0.997889\tsystem UNK || nonlinear UNK system\tREL\tmethod used for task\n", "0.997889\tsystem UNK || nonlinear UNK system\tREL\tmethod used for task\n", "0.997819\tsubtractive clustering || genetic algorithm\tREL\tmethod used for task\n", "0.997750\thybrid algorithm || UNK swarm optimization\tREL\tmethod used for task\n", "0.997748\tUNK swarm optimization || search algorithm\tREL\tmethod used for task\n", "0.997748\tUNK search algorithm || UNK swarm optimization\tREL\tmethod used for task\n", "0.997666\tUNK UNK || text mining system\tREL\tmethod used for task\n", "0.997666\tUNK UNK UNK || text mining system\tREL\tmethod used for task\n", "0.997547\toptimal UNK || hybrid optimization method\tREL\tmethod used for task\n", "0.997466\ttruss structures || UNK problem\tREL\tmethod used for task\n", "0.997466\ttruss structures || UNK problem\tREL\tmethod used for task\n", "0.997419\thybrid method || genetic algorithm\tREL\tmethod used for task\n", "0.997419\tgenetic algorithm || hybrid method\tREL\tmethod used for task\n", "0.997419\thybrid method || genetic algorithm\tREL\tmethod used for task\n", "0.997309\tgenetic algorithm || local search\tREL\tmethod used for task\n", "0.997239\tUNK system || genetic algorithm\tREL\tmethod used for task\n", "0.997239\tgenetic algorithm || UNK system\tREL\tmethod used for task\n", "0.997214\tsimulated annealing || optimization\tREL\tmethod used for task\n", "0.997214\toptimization UNK || simulated annealing\tREL\tmethod used for task\n", "0.997015\toptimization problem || genetic\tREL\tmethod used for task\n", "0.997015\toptimization problem || genetic\tREL\tmethod used for task\n", "0.997000\tUNK UNK || rough set theory\tREL\tmethod used for task\n", "0.997000\trough set theory || UNK\tREL\tmethod used for task\n", "0.997000\trough set theory || UNK\tREL\tmethod used for task\n", "0.996904\tUNK structures || data structures\tREL\tmethod used for task\n", "0.996894\ttext mining || neural networks\tREL\tmethod used for task\n", "0.996837\tUNK swarm optimization || local search\tREL\tmethod used for task\n", "0.996754\tUNK swarm optimization || UNK UNK system\tREL\tmethod used for task\n", "0.996754\tswarm optimization || UNK UNK UNK system\tREL\tmethod used for task\n", "0.996653\toptimization problem || UNK problem\tREL\tmethod used for task\n", "0.996566\toptimization UNK || UNK system design\tREL\tmethod used for task\n", "0.996432\tdata clustering || optimization algorithm\tREL\tmethod used for task\n", "0.996311\tgenetic algorithm || optimization\tREL\tmethod used for task\n", "0.996311\tgenetic algorithm || optimization\tREL\tmethod used for task\n", "0.996240\tsystem performance || UNK system\tREL\tmethod used for task\n", "0.996137\tUNK search algorithm || local search\tREL\tmethod used for task\n", "0.995991\tUNK set || UNK UNK system\tREL\tmethod used for task\n", "0.995976\tUNK structures || local search\tREL\tmethod used for task\n", "0.995870\tsimulated annealing || UNK problem\tREL\tmethod used for task\n", "0.995865\tUNK UNK algorithm || UNK optimization problem\tREL\tmethod used for task\n", "0.995865\tUNK algorithm || optimization problem\tREL\tmethod used for task\n", "0.995865\tUNK algorithm || optimization problem\tREL\tmethod used for task\n", "0.995865\tUNK algorithm || optimization problem\tREL\tmethod used for task\n", "0.995757\tclustering algorithm || clustering\tREL\tmethod used for task\n", "0.995754\tgenetic algorithm || ensemble\tREL\tmethod used for task\n", "0.995715\tUNK search || optimization problem\tREL\tmethod used for task\n", "0.995715\toptimization problem || UNK search\tREL\tmethod used for task\n", "0.995715\tUNK search || optimization problem\tREL\tmethod used for task\n", "0.995687\tneural UNK system || UNK UNK system\tREL\tmethod used for task\n", "0.995548\tartificial neural networks || genetic algorithm\tREL\tmethod used for task\n", "0.995402\tsurrogate model || UNK swarm algorithm\tREL\tmethod used for task\n", "0.995058\tsystem UNK || UNK model building\tREL\tmethod used for task\n", "0.995026\tUNK model || supply chain problem\tREL\tmethod used for task\n", "0.994978\tUNK optimization || system performance\tREL\tmethod used for task\n", "0.994898\tUNK algorithm || simulated annealing\tREL\tmethod used for task\n", "0.994898\tUNK algorithm || simulated annealing\tREL\tmethod used for task\n", "0.994898\tsimulated annealing || UNK algorithm\tREL\tmethod used for task\n", "0.994898\tsimulated annealing || UNK algorithm\tREL\tmethod used for task\n", "0.994853\tUNK design || truss structures\tREL\tmethod used for task\n", "0.994713\tsimulated annealing || UNK search\tREL\tmethod used for task\n", "0.994713\tsimulated annealing || UNK search\tREL\tmethod used for task\n", "0.994707\tUNK optimization || search algorithm\tREL\tmethod used for task\n", "0.994707\tUNK search algorithm || UNK optimization\tREL\tmethod used for task\n", "0.994579\tlocal search || local search\tREL\tmethod used for task\n", "0.994533\tgenetic algorithm || UNK UNK problem\tREL\tmethod used for task\n", "0.994533\tgenetic algorithm || UNK problem\tREL\tmethod used for task\n", "0.994533\tUNK problem || genetic algorithm\tREL\tmethod used for task\n", "0.994533\tgenetic algorithm || UNK UNK UNK problem\tREL\tmethod used for task\n", "0.994442\tUNK set theory || UNK method\tREL\tmethod used for task\n", "0.994292\tUNK system || system\tREL\tmethod used for task\n", "0.994292\tUNK system || UNK system\tREL\tmethod used for task\n", "0.993990\ttopology optimization || UNK UNK method\tREL\tmethod used for task\n", "0.993703\tclustering algorithm || UNK algorithm\tREL\tmethod used for task\n", "0.993511\toptimal UNK UNK || nonlinear system\tREL\tmethod used for task\n", "0.993401\tUNK set || UNK clustering\tREL\tmethod used for task\n", "0.993249\tgenetic algorithm || UNK algorithm\tREL\tmethod used for task\n", "0.993249\tUNK algorithm || genetic algorithm\tREL\tmethod used for task\n", "0.993249\tUNK algorithm || genetic algorithm\tREL\tmethod used for task\n", "0.993249\tUNK algorithm || genetic algorithm\tREL\tmethod used for task\n", "0.993248\tagent-based model || supply chain\tREL\tmethod used for task\n", "0.993129\tclustering method || UNK UNK problem\tREL\tmethod used for task\n", "0.993035\tUNK algorithm || inverse problem\tREL\tmethod used for task\n", "0.993013\thybrid genetic algorithm || UNK\tREL\tmethod used for task\n", "0.993013\thybrid genetic algorithm || UNK\tREL\tmethod used for task\n", "0.993013\thybrid algorithm || genetic\tREL\tmethod used for task\n", "0.993005\tgenetic algorithm || UNK search\tREL\tmethod used for task\n", "0.993005\tgenetic algorithm || UNK search\tREL\tmethod used for task\n", "0.993005\tgenetic algorithm || UNK search\tREL\tmethod used for task\n", "0.993005\tgenetic algorithm || UNK search\tREL\tmethod used for task\n", "0.993005\tgenetic algorithm || UNK search\tREL\tmethod used for task\n", "0.992876\thybrid method || optimization\tREL\tmethod used for task\n", "0.992871\trough set || UNK\tREL\tmethod used for task\n", "0.992413\tneural networks || UNK system\tREL\tmethod used for task\n", "0.992382\tUNK system || optimization\tREL\tmethod used for task\n", "0.992382\tUNK system || UNK optimization\tREL\tmethod used for task\n", "0.992336\ttext mining || UNK techniques\tREL\tmethod used for task\n", "0.992318\tUNK model || partition problem\tREL\tmethod used for task\n", "0.992162\tUNK search algorithm || UNK problem\tREL\tmethod used for task\n", "0.992109\toptimal design || UNK system\tREL\tmethod used for task\n", "0.992109\toptimal design || UNK system\tREL\tmethod used for task\n", "0.992074\tUNK problem || UNK UNK set\tREL\tmethod used for task\n", "0.992069\tUNK UNK algorithm || UNK swarm optimization\tREL\tmethod used for task\n", "0.992069\tUNK swarm optimization || UNK algorithm\tREL\tmethod used for task\n", "0.992069\tUNK swarm optimization || UNK algorithm\tREL\tmethod used for task\n", "0.992069\tUNK swarm optimization algorithm || UNK\tREL\tmethod used for task\n", "0.992069\tUNK swarm optimization || UNK algorithm\tREL\tmethod used for task\n", "0.992069\tUNK swarm optimization || UNK algorithm\tREL\tmethod used for task\n", "0.992069\tUNK swarm optimization algorithm || UNK\tREL\tmethod used for task\n", "0.991836\tUNK structures || problem\tREL\tmethod used for task\n", "0.991792\thybrid UNK || UNK swarm optimization\tREL\tmethod used for task\n", "0.991783\tUNK swarm optimization || UNK search\tREL\tmethod used for task\n", "0.991783\tUNK swarm optimization || UNK search\tREL\tmethod used for task\n", "0.991783\tUNK swarm optimization || UNK search\tREL\tmethod used for task\n", "0.991669\tUNK detection || UNK swarm optimization method\tREL\tmethod used for task\n", "0.991613\toptimization algorithm || UNK UNK design\tREL\tmethod used for task\n", "0.991525\tgenetic algorithm || optimal\tREL\tmethod used for task\n", "0.991486\tnonlinear models || neural networks\tREL\tmethod used for task\n", "0.991470\tUNK model || genetic algorithm\tREL\tmethod used for task\n", "0.991470\tUNK genetic algorithm || model\tREL\tmethod used for task\n", "0.990908\tgenetic algorithm || UNK method\tREL\tmethod used for task\n", "0.990749\tshell structures || UNK\tREL\tmethod used for task\n", "0.990749\tshell structures || UNK\tREL\tmethod used for task\n", "0.990749\tshell structures || UNK\tREL\tmethod used for task\n", "0.990730\toptimization method || UNK UNK method\tREL\tmethod used for task\n", "0.990630\toptimization techniques || optimization technique\tREL\tmethod used for task\n", "0.990619\thybrid method || UNK networks\tREL\tmethod used for task\n", "0.990610\tpso algorithm || UNK UNK problem\tREL\tmethod used for task\n", "0.990580\tgenetic search || UNK method\tREL\tmethod used for task\n", "0.990537\tgenetic algorithm || local\tREL\tmethod used for task\n", "0.990520\toptimization model || optimal shape\tREL\tmethod used for task\n", "0.990505\tUNK problem || optimal algorithm\tREL\tmethod used for task\n", "0.990326\tsearch algorithm || UNK algorithm\tREL\tmethod used for task\n", "0.990269\tdata clustering || UNK system\tREL\tmethod used for task\n", "0.990107\tUNK set theory || UNK performance\tREL\tmethod used for task\n", "0.989988\thybrid algorithm || search\tREL\tmethod used for task\n", "0.989988\thybrid algorithm || UNK search\tREL\tmethod used for task\n", "0.989988\thybrid algorithm || UNK search\tREL\tmethod used for task\n", "0.989988\tUNK search || hybrid algorithm\tREL\tmethod used for task\n", "0.989983\tmodel UNK || UNK swarm optimization\tREL\tmethod used for task\n", "0.989977\tUNK search || search algorithm\tREL\tmethod used for task\n", "0.989977\tUNK search algorithm || UNK search\tREL\tmethod used for task\n", "0.989977\tUNK search algorithm || UNK search\tREL\tmethod used for task\n", "0.989977\tUNK search algorithm || UNK search\tREL\tmethod used for task\n", "0.989940\tgenetic UNK system || UNK\tREL\tmethod used for task\n", "0.989924\tUNK algorithm || UNK structures\tREL\tmethod used for task\n", "0.989904\tgenetic optimization || model parameters\tREL\tmethod used for task\n", "0.989864\tUNK search || UNK set\tREL\tmethod used for task\n", "0.989761\thybrid technique || hybrid system\tREL\tmethod used for task\n", "0.989323\tUNK UNK swarm optimization || UNK UNK method\tREL\tmethod used for task\n", "0.989323\tUNK swarm optimization || UNK UNK method\tREL\tmethod used for task\n", "0.989194\tdynamic UNK algorithm || optimal\tREL\tmethod used for task\n", "0.989194\tdynamic UNK algorithm || optimal\tREL\tmethod used for task\n", "0.988888\tUNK swarm optimization || local\tREL\tmethod used for task\n", "0.988732\tproblem UNK || system\tREL\tmethod used for task\n", "0.988732\tUNK system || UNK problem\tREL\tmethod used for task\n", "0.988732\tUNK problem || system\tREL\tmethod used for task\n", "0.988604\tUNK UNK || truss structures\tREL\tmethod used for task\n", "0.988415\tpso algorithm || UNK algorithm\tREL\tmethod used for task\n", "0.987760\toptimization algorithm || UNK performance\tREL\tmethod used for task\n", "0.987760\toptimization algorithm || UNK performance\tREL\tmethod used for task\n", "0.987585\tmodeling language || domain model\tREL\tmethod used for task\n", "0.987281\tUNK UNK || model structures\tREL\tmethod used for task\n", "0.987030\tdata clustering || optimization\tREL\tmethod used for task\n", "0.987030\tdata clustering || optimization\tREL\tmethod used for task\n", "0.986999\thybrid algorithm || UNK method\tREL\tmethod used for task\n", "0.986984\tUNK search algorithm || UNK method\tREL\tmethod used for task\n", "0.986924\tswarm intelligence || UNK optimization\tREL\tmethod used for task\n", "0.986839\tUNK set || UNK UNK method\tREL\tmethod used for task\n", "0.986814\tnonlinear UNK || UNK set\tREL\tmethod used for task\n", "0.986592\tgenetic optimization || UNK\tREL\tmethod used for task\n", "0.986470\thybrid algorithm || local\tREL\tmethod used for task\n", "0.986455\tUNK search algorithm || local\tREL\tmethod used for task\n", "0.986455\tUNK algorithm || local search\tREL\tmethod used for task\n", "0.986455\tUNK algorithm || local search\tREL\tmethod used for task\n", "0.986455\tlocal search || UNK algorithm\tREL\tmethod used for task\n", "0.986455\tUNK algorithm || local search\tREL\tmethod used for task\n", "0.986455\tlocal search || UNK algorithm\tREL\tmethod used for task\n", "0.986104\tUNK algorithm || UNK system\tREL\tmethod used for task\n", "0.986104\tUNK system || UNK algorithm\tREL\tmethod used for task\n", "0.986104\tUNK algorithm || system\tREL\tmethod used for task\n", "0.985968\tUNK search || local search\tREL\tmethod used for task\n", "0.985968\tlocal search || UNK search\tREL\tmethod used for task\n", "0.985968\tlocal search || search\tREL\tmethod used for task\n", "0.985621\tUNK UNK || hybrid system\tREL\tmethod used for task\n", "0.985621\thybrid system || UNK\tREL\tmethod used for task\n", "0.985621\thybrid UNK || UNK UNK system\tREL\tmethod used for task\n", "0.985621\thybrid system || UNK\tREL\tmethod used for task\n", "0.985621\thybrid system || UNK\tREL\tmethod used for task\n", "0.985621\thybrid system || UNK\tREL\tmethod used for task\n", "0.985621\tUNK UNK || hybrid system\tREL\tmethod used for task\n", "0.985605\tUNK search || UNK system\tREL\tmethod used for task\n", "0.985565\ttopology UNK || wireless sensor networks\tREL\tmethod used for task\n", "0.985474\toptimal UNK || pso algorithm\tREL\tmethod used for task\n", "0.985368\tUNK domain || UNK system\tREL\tmethod used for task\n", "0.985219\toptimal algorithm || UNK model\tREL\tmethod used for task\n", "0.985091\tUNK model || solid mechanics\tREL\tmethod used for task\n", "0.985031\tUNK set theory || UNK\tREL\tmethod used for task\n", "0.985031\tUNK set theory || UNK\tREL\tmethod used for task\n", "0.985031\tUNK set theory || UNK\tREL\tmethod used for task\n", "0.985031\tUNK UNK || UNK set theory\tREL\tmethod used for task\n", "0.984989\toptimization UNK || UNK UNK UNK problem\tREL\tmethod used for task\n", "0.984989\tUNK UNK || optimization problem\tREL\tmethod used for task\n", "0.984989\tUNK UNK UNK || optimization problem\tREL\tmethod used for task\n", "0.984989\tUNK optimization problem || UNK\tREL\tmethod used for task\n", "0.984989\tUNK UNK || UNK optimization problem\tREL\tmethod used for task\n", "0.984989\toptimization problem || UNK\tREL\tmethod used for task\n", "0.984989\toptimization problem || UNK\tREL\tmethod used for task\n", "0.984989\toptimization problem || UNK\tREL\tmethod used for task\n", "0.984989\toptimization problem || UNK\tREL\tmethod used for task\n", "0.984989\tUNK UNK || optimization problem\tREL\tmethod used for task\n", "0.984989\toptimization problem || UNK\tREL\tmethod used for task\n", "0.984989\tUNK UNK || UNK optimization problem\tREL\tmethod used for task\n", "0.984989\toptimization problem || UNK\tREL\tmethod used for task\n", "0.984989\tUNK UNK || UNK optimization problem\tREL\tmethod used for task\n", "0.984943\tUNK clustering || data mining\tREL\tmethod used for task\n", "0.984926\tUNK clustering || artificial neural networks\tREL\tmethod used for task\n", "0.984500\tsearch space || optimization problem\tREL\tmethod used for task\n", "0.984339\tdomain model || UNK model\tREL\tmethod used for task\n", "0.984339\tdomain model || UNK model\tREL\tmethod used for task\n", "0.984121\tUNK algorithm || nonlinear model\tREL\tmethod used for task\n", "0.983828\ttopology optimization || UNK\tREL\tmethod used for task\n", "0.983828\ttopology optimization || UNK\tREL\tmethod used for task\n", "0.983486\tUNK algorithm || wireless sensor networks\tREL\tmethod used for task\n", "0.983486\tUNK algorithm || wireless sensor networks\tREL\tmethod used for task\n", "0.983449\tUNK UNK simulation || hybrid genetic algorithm\tREL\tmethod used for task\n", "0.982929\tUNK optimization || dynamic\tREL\tmethod used for task\n", "0.982929\tUNK UNK UNK || dynamic optimization\tREL\tmethod used for task\n", "0.982917\tmodel UNK || local search\tREL\tmethod used for task\n", "0.982917\tlocal search || UNK model\tREL\tmethod used for task\n", "0.982803\tUNK UNK models || UNK system\tREL\tmethod used for task\n", "0.982476\tUNK model || system\tREL\tmethod used for task\n", "0.982476\tmodel UNK || system\tREL\tmethod used for task\n", "0.982425\tUNK UNK || wireless UNK UNK networks\tREL\tmethod used for task\n", "0.982425\tUNK UNK || wireless UNK networks\tREL\tmethod used for task\n", "0.982097\tsystem UNK || supply\tREL\tmethod used for task\n", "0.981526\tUNK UNK || simulated annealing\tREL\tmethod used for task\n", "0.981526\tsimulated annealing || UNK\tREL\tmethod used for task\n", "0.981526\tsimulated annealing || UNK\tREL\tmethod used for task\n", "0.981503\tUNK algorithm || optimization\tREL\tmethod used for task\n", "0.981503\tUNK algorithm || UNK optimization\tREL\tmethod used for task\n", "0.981503\tUNK algorithm || optimization\tREL\tmethod used for task\n", "0.981503\toptimization algorithm || UNK\tREL\tmethod used for task\n", "0.981503\tUNK algorithm || UNK optimization\tREL\tmethod used for task\n", "0.981503\tUNK UNK algorithm || UNK optimization\tREL\tmethod used for task\n", "0.981503\tUNK UNK algorithm || UNK optimization\tREL\tmethod used for task\n", "0.981503\tUNK UNK algorithm || UNK optimization\tREL\tmethod used for task\n", "0.981503\tUNK UNK algorithm || UNK optimization\tREL\tmethod used for task\n", "0.981503\tUNK UNK algorithm || UNK optimization\tREL\tmethod used for task\n", "0.981503\tUNK UNK || UNK UNK optimization algorithm\tREL\tmethod used for task\n", "0.981503\tUNK optimization || UNK algorithm\tREL\tmethod used for task\n", "0.981503\tUNK algorithm || UNK optimization\tREL\tmethod used for task\n", "0.981503\tUNK UNK || UNK optimization algorithm\tREL\tmethod used for task\n", "0.981503\tUNK algorithm || optimization\tREL\tmethod used for task\n", "0.981503\tUNK optimization || UNK algorithm\tREL\tmethod used for task\n", "0.981503\tUNK algorithm || optimization\tREL\tmethod used for task\n", "0.981398\tUNK system || UNK UNK techniques\tREL\tmethod used for task\n", "0.981331\tUNK system || UNK method\tREL\tmethod used for task\n", "0.981331\tUNK system || UNK method\tREL\tmethod used for task\n", "0.981331\tUNK system || UNK UNK method\tREL\tmethod used for task\n", "0.980941\tneural networks || hybrid\tREL\tmethod used for task\n", "0.980842\tUNK optimization || search\tREL\tmethod used for task\n", "0.980842\tUNK search || UNK optimization\tREL\tmethod used for task\n", "0.980842\toptimization UNK || UNK search\tREL\tmethod used for task\n", "0.980530\tdomain UNK || UNK optimization\tREL\tmethod used for task\n", "0.980279\tsubtractive clustering || experimental data\tREL\tmethod used for task\n", "0.980225\tgenetic UNK || UNK UNK problem\tREL\tmethod used for task\n", "0.980072\tUNK problem || rough\tREL\tmethod used for task\n", "0.979881\tUNK swarm || search method\tREL\tmethod used for task\n", "0.979360\tUNK UNK || text mining\tREL\tmethod used for task\n", "0.978798\tUNK analysis || shell structures\tREL\tmethod used for task\n", "0.978798\tshell structures || UNK analysis\tREL\tmethod used for task\n", "0.978457\tnumerical UNK || shell structures\tREL\tmethod used for task\n", "0.978227\tUNK models || supply chain\tREL\tmethod used for task\n", "0.978227\tUNK models || supply chain\tREL\tmethod used for task\n", "0.978222\ttruss structures || computational\tREL\tmethod used for task\n", "0.978222\ttruss structures || computational\tREL\tmethod used for task\n", "0.977914\twireless sensor networks || UNK UNK techniques\tREL\tmethod used for task\n", "0.977877\tUNK UNK problem || UNK problem\tREL\tmethod used for task\n", "0.977814\tUNK model || supply chain\tREL\tmethod used for task\n", "0.977814\tUNK model || supply chain\tREL\tmethod used for task\n", "0.977814\tsupply chain || UNK model\tREL\tmethod used for task\n", "0.977313\tsystem UNK || design\tREL\tmethod used for task\n", "0.977313\tsystem design || UNK\tREL\tmethod used for task\n", "0.977313\tsystem design || UNK\tREL\tmethod used for task\n", "0.977313\tsystem design || UNK\tREL\tmethod used for task\n", "0.977269\tUNK clustering || UNK algorithm\tREL\tmethod used for task\n", "0.977269\tUNK UNK || clustering algorithm\tREL\tmethod used for task\n", "0.977269\tclustering algorithm || UNK\tREL\tmethod used for task\n", "0.977244\tUNK system || data mining technique\tREL\tmethod used for task\n", "0.977136\toptimization models || UNK\tREL\tmethod used for task\n", "0.976703\toptimization model || UNK\tREL\tmethod used for task\n", "0.976703\tUNK UNK optimization || UNK UNK model\tREL\tmethod used for task\n", "0.976703\toptimization model || UNK\tREL\tmethod used for task\n", "0.976703\toptimization model || UNK\tREL\tmethod used for task\n", "0.976703\toptimization model || UNK\tREL\tmethod used for task\n", "0.976703\toptimization model || UNK\tREL\tmethod used for task\n", "0.976703\tUNK model || optimization\tREL\tmethod used for task\n", "0.976703\tUNK UNK || UNK optimization model\tREL\tmethod used for task\n", "0.976548\toptimization problem || solution method\tREL\tmethod used for task\n", "0.976446\tclustering algorithm || data\tREL\tmethod used for task\n", "0.976212\tUNK UNK || concrete structures\tREL\tmethod used for task\n", "0.976202\tsupply UNK || optimization\tREL\tmethod used for task\n", "0.975661\tUNK UNK UNK || genetic algorithm\tREL\tmethod used for task\n", "0.975661\tgenetic algorithm || UNK\tREL\tmethod used for task\n", "0.975661\tgenetic algorithm || UNK\tREL\tmethod used for task\n", "0.975661\tUNK UNK || genetic algorithm\tREL\tmethod used for task\n", "0.975661\tUNK algorithm || genetic\tREL\tmethod used for task\n", "0.975661\tUNK algorithm || genetic\tREL\tmethod used for task\n", "0.975661\tgenetic algorithm || UNK\tREL\tmethod used for task\n", "0.975661\tUNK UNK || genetic algorithm\tREL\tmethod used for task\n", "0.975661\tUNK UNK || genetic algorithm\tREL\tmethod used for task\n", "0.975661\tUNK UNK || genetic algorithm\tREL\tmethod used for task\n", "0.975661\tgenetic algorithm || UNK\tREL\tmethod used for task\n", "0.975661\tgenetic algorithm || UNK\tREL\tmethod used for task\n", "0.975661\tgenetic algorithm || UNK\tREL\tmethod used for task\n", "0.975661\tgenetic algorithm || UNK\tREL\tmethod used for task\n", "0.975661\tgenetic algorithm || UNK\tREL\tmethod used for task\n", "0.975661\tUNK UNK || genetic algorithm\tREL\tmethod used for task\n", "0.975661\tUNK UNK || genetic algorithm\tREL\tmethod used for task\n", "0.975661\tgenetic algorithm || UNK\tREL\tmethod used for task\n", "0.975661\tUNK UNK || genetic algorithm\tREL\tmethod used for task\n", "0.975661\tgenetic algorithm || UNK\tREL\tmethod used for task\n", "0.975661\tgenetic algorithm || UNK\tREL\tmethod used for task\n", "0.975661\tUNK UNK || genetic algorithm\tREL\tmethod used for task\n", "0.975661\tUNK UNK || genetic algorithm\tREL\tmethod used for task\n", "0.975661\tgenetic algorithm || UNK\tREL\tmethod used for task\n", "0.975661\tgenetic algorithm || UNK\tREL\tmethod used for task\n", "0.975661\tgenetic algorithm || UNK\tREL\tmethod used for task\n", "0.975661\tgenetic algorithm || UNK\tREL\tmethod used for task\n", "0.975661\tgenetic algorithm || UNK\tREL\tmethod used for task\n", "0.975661\tgenetic algorithm || UNK\tREL\tmethod used for task\n", "0.975598\tsupply chain || UNK modeling\tREL\tmethod used for task\n", "0.975378\tneural networks || UNK techniques\tREL\tmethod used for task\n", "0.975191\toptimization method || UNK\tREL\tmethod used for task\n", "0.975191\tUNK method || UNK optimization\tREL\tmethod used for task\n", "0.975191\toptimization UNK || UNK method\tREL\tmethod used for task\n", "0.975191\toptimization UNK || UNK method\tREL\tmethod used for task\n", "0.975145\tUNK optimization || UNK UNK nonlinear\tREL\tmethod used for task\n", "0.975145\tnonlinear optimization || UNK\tREL\tmethod used for task\n", "0.974866\tdynamic UNK || UNK problem\tREL\tmethod used for task\n", "0.974797\tUNK search || genetic\tREL\tmethod used for task\n", "0.973789\tartificial neural network || hybrid genetic algorithm\tREL\tmethod used for task\n", "0.973763\tensemble UNK || UNK models\tREL\tmethod used for task\n", "0.973236\tensemble classifier || UNK\tREL\tmethod used for task\n", "0.973236\tUNK UNK UNK || classifier ensemble\tREL\tmethod used for task\n", "0.972786\tUNK UNK algorithm || UNK UNK problem\tREL\tmethod used for task\n", "0.972786\tUNK algorithm || UNK problem\tREL\tmethod used for task\n", "0.972786\tUNK algorithm || UNK problem\tREL\tmethod used for task\n", "0.972786\tUNK UNK algorithm || UNK problem\tREL\tmethod used for task\n", "0.972786\tUNK algorithm || UNK problem\tREL\tmethod used for task\n", "0.972786\tUNK algorithm || problem\tREL\tmethod used for task\n", "0.972786\tUNK UNK problem || UNK algorithm\tREL\tmethod used for task\n", "0.972786\tUNK algorithm || UNK UNK problem\tREL\tmethod used for task\n", "0.972786\tUNK problem || UNK UNK algorithm\tREL\tmethod used for task\n", "0.972786\tUNK algorithm || UNK problem\tREL\tmethod used for task\n", "0.972786\tUNK algorithm || UNK problem\tREL\tmethod used for task\n", "0.972786\tUNK problem || UNK algorithm\tREL\tmethod used for task\n", "0.972786\tUNK problem || UNK algorithm\tREL\tmethod used for task\n", "0.972786\tUNK UNK problem || UNK UNK algorithm\tREL\tmethod used for task\n", "0.971996\tartificial neural networks || model\tREL\tmethod used for task\n", "0.971829\tUNK theory || UNK optimization\tREL\tmethod used for task\n", "0.971823\tUNK problem || UNK search\tREL\tmethod used for task\n", "0.971823\tUNK UNK search || UNK problem\tREL\tmethod used for task\n", "0.971823\tUNK UNK UNK search || UNK UNK problem\tREL\tmethod used for task\n", "0.971713\tshell UNK || shell\tREL\tmethod used for task\n", "0.971585\tdynamic UNK || UNK shell\tREL\tmethod used for task\n", "0.971540\tUNK UNK || ensemble method\tREL\tmethod used for task\n", "0.971495\tUNK swarm optimization || UNK\tREL\tmethod used for task\n", "0.971495\tUNK UNK || UNK swarm optimization\tREL\tmethod used for task\n", "0.971495\tUNK UNK swarm optimization || UNK\tREL\tmethod used for task\n", "0.971495\tUNK swarm optimization || UNK\tREL\tmethod used for task\n", "0.971495\tUNK swarm optimization || UNK\tREL\tmethod used for task\n", "0.971495\tUNK swarm optimization || UNK\tREL\tmethod used for task\n", "0.971495\tUNK swarm optimization || UNK\tREL\tmethod used for task\n", "0.971495\tUNK UNK || UNK swarm optimization\tREL\tmethod used for task\n", "0.971495\tUNK swarm optimization || UNK\tREL\tmethod used for task\n", "0.971495\tUNK swarm optimization || UNK\tREL\tmethod used for task\n", "0.971495\tUNK UNK || UNK swarm optimization\tREL\tmethod used for task\n", "0.971495\tUNK UNK || UNK swarm optimization\tREL\tmethod used for task\n", "0.971495\tUNK UNK || UNK UNK swarm optimization\tREL\tmethod used for task\n", "0.971495\tswarm optimization || UNK\tREL\tmethod used for task\n", "0.971410\toptimization problem || computational\tREL\tmethod used for task\n", "0.971402\tUNK clustering || UNK model\tREL\tmethod used for task\n", "0.971367\tdomain UNK || problem\tREL\tmethod used for task\n", "0.970580\tUNK UNK swarm optimization || search space\tREL\tmethod used for task\n", "0.970043\tUNK models || UNK networks\tREL\tmethod used for task\n", "0.969664\tclustering techniques || UNK\tREL\tmethod used for task\n", "0.969557\tclustering method || UNK\tREL\tmethod used for task\n", "0.969557\tUNK UNK UNK UNK || clustering method\tREL\tmethod used for task\n", "0.969557\tclustering method || UNK\tREL\tmethod used for task\n", "0.969557\tclustering method || UNK\tREL\tmethod used for task\n", "0.969557\tUNK clustering || UNK method\tREL\tmethod used for task\n", "0.969480\tUNK model || UNK networks\tREL\tmethod used for task\n", "0.969480\tUNK networks || UNK model\tREL\tmethod used for task\n", "0.969339\thybrid model || linear model\tREL\tmethod used for task\n", "0.969104\tdynamic UNK || UNK algorithm\tREL\tmethod used for task\n", "0.968725\tpso algorithm || model parameters\tREL\tmethod used for task\n", "0.968100\thybrid UNK || inverse\tREL\tmethod used for task\n", "0.968050\thybrid UNK || dynamic\tREL\tmethod used for task\n", "0.967422\tgenetic UNK || UNK UNK method\tREL\tmethod used for task\n", "0.967422\tgenetic UNK || UNK UNK method\tREL\tmethod used for task\n", "0.967422\tgenetic UNK || UNK method\tREL\tmethod used for task\n", "0.967110\tUNK UNK || system performance\tREL\tmethod used for task\n", "0.967110\tUNK performance || system\tREL\tmethod used for task\n", "0.967110\tUNK UNK || system performance\tREL\tmethod used for task\n", "0.967110\tUNK system || UNK performance\tREL\tmethod used for task\n", "0.967110\tUNK UNK system || UNK performance\tREL\tmethod used for task\n", "0.967110\tUNK UNK system || UNK performance\tREL\tmethod used for task\n", "0.967110\tsystem performance || UNK\tREL\tmethod used for task\n", "0.966872\tUNK UNK || strip method\tREL\tmethod used for task\n", "0.966853\tgenetic UNK systems || genetic\tREL\tmethod used for task\n", "0.966563\tUNK UNK algorithm || UNK algorithm\tREL\tmethod used for task\n", "0.966563\tUNK UNK algorithm || UNK algorithm\tREL\tmethod used for task\n", "0.966563\tUNK UNK algorithm || UNK algorithm\tREL\tmethod used for task\n", "0.966563\tUNK UNK algorithm || UNK algorithm\tREL\tmethod used for task\n", "0.966563\tUNK algorithm || UNK UNK algorithm\tREL\tmethod used for task\n", "0.966563\tUNK algorithm || UNK algorithm\tREL\tmethod used for task\n", "0.966563\tUNK algorithm || UNK algorithm\tREL\tmethod used for task\n", "0.966563\tUNK algorithm || UNK algorithm\tREL\tmethod used for task\n", "0.966563\tUNK algorithm || UNK algorithm\tREL\tmethod used for task\n", "0.966014\toptimal UNK || UNK problem\tREL\tmethod used for task\n", "0.966014\toptimal UNK || UNK problem\tREL\tmethod used for task\n", "0.965947\tUNK algorithm || UNK mechanics\tREL\tmethod used for task\n", "0.965802\tUNK UNK model || UNK UNK problem\tREL\tmethod used for task\n", "0.965802\tUNK model || UNK UNK problem\tREL\tmethod used for task\n", "0.965802\tUNK UNK problem || UNK model\tREL\tmethod used for task\n", "0.965802\tUNK model || UNK UNK problem\tREL\tmethod used for task\n", "0.965802\tUNK model || UNK problem\tREL\tmethod used for task\n", "0.965426\thybrid algorithm || UNK\tREL\tmethod used for task\n", "0.965426\thybrid algorithm || UNK\tREL\tmethod used for task\n", "0.965426\tUNK UNK || hybrid algorithm\tREL\tmethod used for task\n", "0.965426\thybrid algorithm || UNK\tREL\tmethod used for task\n", "0.965426\thybrid algorithm || UNK\tREL\tmethod used for task\n", "0.965426\thybrid algorithm || UNK\tREL\tmethod used for task\n", "0.965426\tUNK UNK || hybrid algorithm\tREL\tmethod used for task\n", "0.965387\tUNK UNK || search algorithm\tREL\tmethod used for task\n", "0.965387\tUNK UNK || UNK search algorithm\tREL\tmethod used for task\n", "0.965387\tUNK UNK || UNK search algorithm\tREL\tmethod used for task\n", "0.965387\tUNK search || UNK algorithm\tREL\tmethod used for task\n", "0.965387\tUNK UNK || UNK UNK search algorithm\tREL\tmethod used for task\n", "0.965387\tUNK search algorithm || UNK\tREL\tmethod used for task\n", "0.965387\tUNK algorithm || search\tREL\tmethod used for task\n", "0.965387\tUNK search algorithm || UNK\tREL\tmethod used for task\n", "0.965387\tsearch algorithm || UNK\tREL\tmethod used for task\n", "0.965387\tUNK search algorithm || UNK\tREL\tmethod used for task\n", "0.965387\tUNK search algorithm || UNK\tREL\tmethod used for task\n", "0.965387\tUNK search algorithm || UNK\tREL\tmethod used for task\n", "0.965387\tUNK search algorithm || UNK\tREL\tmethod used for task\n", "0.965387\tUNK search algorithm || UNK\tREL\tmethod used for task\n", "0.965387\tUNK algorithm || UNK search\tREL\tmethod used for task\n", "0.965387\tsearch algorithm || UNK\tREL\tmethod used for task\n", "0.965387\tUNK search algorithm || UNK\tREL\tmethod used for task\n", "0.965387\tsearch algorithm || UNK\tREL\tmethod used for task\n", "0.965310\tnumerical UNK || UNK optimization problem\tREL\tmethod used for task\n", "0.965145\tUNK UNK mining || design\tREL\tmethod used for task\n", "0.965008\tUNK set || UNK\tREL\tmethod used for task\n", "0.965008\tUNK UNK || UNK set\tREL\tmethod used for task\n", "0.965008\tUNK set || UNK\tREL\tmethod used for task\n", "0.965008\tUNK set || UNK\tREL\tmethod used for task\n", "0.965008\tUNK UNK set || UNK\tREL\tmethod used for task\n", "0.965008\tUNK UNK || UNK set\tREL\tmethod used for task\n", "0.965008\tUNK set || UNK\tREL\tmethod used for task\n", "0.965008\tUNK set || UNK\tREL\tmethod used for task\n", "0.965008\tUNK UNK UNK || UNK set\tREL\tmethod used for task\n", "0.965008\tUNK UNK || UNK set\tREL\tmethod used for task\n", "0.965008\tUNK set || UNK\tREL\tmethod used for task\n", "0.965008\tUNK set || UNK\tREL\tmethod used for task\n", "0.965008\tUNK UNK || UNK set\tREL\tmethod used for task\n", "0.965008\tUNK UNK || UNK set\tREL\tmethod used for task\n", "0.965008\tUNK set || UNK\tREL\tmethod used for task\n", "0.964989\thybrid method || model parameters\tREL\tmethod used for task\n", "0.964428\tagent-based modeling || complex networks\tREL\tmethod used for task\n", "0.964225\twind UNK || hybrid optimization method\tREL\tmethod used for task\n", "0.964212\tUNK UNK search || hybrid\tREL\tmethod used for task\n", "0.964172\tsearch UNK || search\tREL\tmethod used for task\n", "0.964172\tUNK search || UNK search\tREL\tmethod used for task\n", "0.964172\tUNK search || search\tREL\tmethod used for task\n", "0.964172\tsearch UNK || UNK search\tREL\tmethod used for task\n", "0.963988\tUNK structures || UNK\tREL\tmethod used for task\n", "0.963757\tUNK UNK || UNK data set\tREL\tmethod used for task\n", "0.963757\tUNK set || UNK data\tREL\tmethod used for task\n", "0.963610\tUNK method || UNK UNK problem\tREL\tmethod used for task\n", "0.962702\tdata structures || UNK\tREL\tmethod used for task\n", "0.962702\tdata structures || UNK\tREL\tmethod used for task\n", "0.962634\tUNK model || system parameters\tREL\tmethod used for task\n", "0.962438\tneural UNK system || UNK\tREL\tmethod used for task\n", "0.962176\tgenetic algorithm || solution method\tREL\tmethod used for task\n", "0.962112\tsystem reliability || UNK\tREL\tmethod used for task\n", "0.962112\tUNK UNK || system reliability\tREL\tmethod used for task\n", "0.962112\tUNK system || UNK reliability\tREL\tmethod used for task\n", "0.961922\tdynamic models || UNK\tREL\tmethod used for task\n", "0.961922\tUNK UNK models || dynamic\tREL\tmethod used for task\n", "0.961567\tUNK UNK || surrogate model\tREL\tmethod used for task\n", "0.961295\tUNK system || UNK detection\tREL\tmethod used for task\n", "0.961295\tUNK system || UNK detection\tREL\tmethod used for task\n", "0.961222\tdomain UNK || UNK ontology\tREL\tmethod used for task\n", "0.961222\tUNK UNK || domain ontology\tREL\tmethod used for task\n", "0.961213\tUNK UNK || dynamic UNK model\tREL\tmethod used for task\n", "0.961213\tUNK UNK || dynamic UNK model\tREL\tmethod used for task\n", "0.961213\tdynamic UNK || UNK model\tREL\tmethod used for task\n", "0.961213\tUNK UNK || dynamic UNK model\tREL\tmethod used for task\n", "0.961091\tdata structures || problem size\tREL\tmethod used for task\n", "0.960764\tsubtractive clustering || UNK\tREL\tmethod used for task\n", "0.960650\tneural network || dynamic system\tREL\tmethod used for task\n", "0.960530\tgenetic UNK || UNK design\tREL\tmethod used for task\n", "0.960530\tUNK design || genetic\tREL\tmethod used for task\n", "0.959610\tUNK UNK model || text\tREL\tmethod used for task\n", "0.959391\tUNK chain model || performance\tREL\tmethod used for task\n", "0.958816\tUNK UNK models || UNK algorithm\tREL\tmethod used for task\n", "0.958816\tUNK UNK algorithm || UNK models\tREL\tmethod used for task\n", "0.958816\tUNK algorithm || UNK models\tREL\tmethod used for task\n", "0.958755\tUNK algorithm || pso\tREL\tmethod used for task\n", "0.958755\tUNK algorithm || pso\tREL\tmethod used for task\n", "0.958755\tUNK UNK || pso algorithm\tREL\tmethod used for task\n", "0.958755\tpso algorithm || UNK\tREL\tmethod used for task\n", "0.958739\tUNK UNK method || dynamic\tREL\tmethod used for task\n", "0.958704\tUNK swarm optimization || optimal solution\tREL\tmethod used for task\n", "0.958309\tUNK UNK || optimal algorithm\tREL\tmethod used for task\n", "0.958309\tUNK UNK || optimal algorithm\tREL\tmethod used for task\n", "0.958308\tUNK algorithm || optimal\tREL\tmethod used for task\n", "0.958051\tUNK algorithm || UNK model\tREL\tmethod used for task\n", "0.958051\tUNK UNK algorithm || UNK model\tREL\tmethod used for task\n", "0.958051\tUNK UNK algorithm || UNK model\tREL\tmethod used for task\n", "0.958051\tUNK algorithm || UNK model\tREL\tmethod used for task\n", "0.958051\tUNK algorithm || UNK model\tREL\tmethod used for task\n", "0.958051\tUNK algorithm || UNK UNK model\tREL\tmethod used for task\n", "0.958051\tUNK algorithm || UNK model\tREL\tmethod used for task\n", "0.958051\tUNK algorithm || UNK model\tREL\tmethod used for task\n", "0.958051\tmodel UNK || UNK algorithm\tREL\tmethod used for task\n", "0.958051\tUNK model || UNK algorithm\tREL\tmethod used for task\n", "0.957427\thybrid models || UNK\tREL\tmethod used for task\n", "0.957427\thybrid models || UNK\tREL\tmethod used for task\n", "0.957427\thybrid models || UNK\tREL\tmethod used for task\n", "0.957427\thybrid models || UNK\tREL\tmethod used for task\n", "0.957166\tUNK model || building\tREL\tmethod used for task\n", "0.957166\tUNK UNK UNK || model building\tREL\tmethod used for task\n", "0.957166\tUNK UNK || UNK model building\tREL\tmethod used for task\n", "0.956904\tlinear models || supply chain\tREL\tmethod used for task\n", "0.956701\tUNK models || domain\tREL\tmethod used for task\n", "0.956638\thybrid model || UNK\tREL\tmethod used for task\n", "0.956638\thybrid model || UNK\tREL\tmethod used for task\n", "0.956638\thybrid model || UNK\tREL\tmethod used for task\n", "0.956573\thybrid system || UNK cost\tREL\tmethod used for task\n", "0.956011\tmodel UNK method || UNK detection\tREL\tmethod used for task\n", "0.955948\tproblem UNK || UNK UNK design\tREL\tmethod used for task\n", "0.955899\tUNK model || UNK domain\tREL\tmethod used for task\n", "0.955899\tUNK model || UNK domain\tREL\tmethod used for task\n", "0.955899\tUNK UNK || domain model\tREL\tmethod used for task\n", "0.955538\tUNK algorithm || UNK UNK techniques\tREL\tmethod used for task\n", "0.955384\tUNK method || UNK algorithm\tREL\tmethod used for task\n", "0.955384\tUNK method || UNK algorithm\tREL\tmethod used for task\n", "0.955384\tUNK method || UNK algorithm\tREL\tmethod used for task\n", "0.955384\tUNK algorithm || UNK UNK method\tREL\tmethod used for task\n", "0.955384\tUNK algorithm || UNK UNK method\tREL\tmethod used for task\n", "0.955384\tUNK algorithm || UNK method\tREL\tmethod used for task\n", "0.955384\tUNK algorithm || UNK method\tREL\tmethod used for task\n", "0.955384\tUNK algorithm || UNK method\tREL\tmethod used for task\n", "0.955384\tUNK UNK method || UNK algorithm\tREL\tmethod used for task\n", "0.955384\tUNK algorithm || UNK method\tREL\tmethod used for task\n", "0.955384\tUNK algorithm || UNK method\tREL\tmethod used for task\n", "0.955360\tUNK networks || detection performance\tREL\tmethod used for task\n", "0.954663\tUNK text || relationship\tREL\tmethod used for task\n", "0.954038\tgenetic algorithm || computational\tREL\tmethod used for task\n", "0.954038\tgenetic algorithm || computational\tREL\tmethod used for task\n", "0.953885\thybrid method || UNK\tREL\tmethod used for task\n", "0.953885\thybrid method || UNK\tREL\tmethod used for task\n", "0.953885\thybrid method || UNK\tREL\tmethod used for task\n", "0.953885\thybrid method || UNK\tREL\tmethod used for task\n", "0.953885\thybrid method || UNK\tREL\tmethod used for task\n", "0.953885\tUNK method || hybrid\tREL\tmethod used for task\n", "0.953885\thybrid method || UNK\tREL\tmethod used for task\n", "0.953834\tUNK UNK || search method\tREL\tmethod used for task\n", "0.953834\tUNK UNK || search method\tREL\tmethod used for task\n", "0.953834\tUNK search || UNK method\tREL\tmethod used for task\n", "0.953834\tUNK search || UNK method\tREL\tmethod used for task\n", "0.953629\tUNK algorithm || local\tREL\tmethod used for task\n", "0.953629\tlocal algorithm || UNK\tREL\tmethod used for task\n", "0.953629\tUNK UNK || local algorithm\tREL\tmethod used for task\n", "0.952954\tUNK neural networks || language\tREL\tmethod used for task\n", "0.952946\tsearch UNK || ontology construction\tREL\tmethod used for task\n", "0.952946\tsearch UNK || ontology construction\tREL\tmethod used for task\n", "0.952256\thybrid method || UNK data\tREL\tmethod used for task\n", "0.952021\tUNK search || local\tREL\tmethod used for task\n", "0.952021\tlocal UNK || UNK search\tREL\tmethod used for task\n", "0.952021\tUNK UNK || local search\tREL\tmethod used for task\n", "0.952021\tlocal UNK || UNK search\tREL\tmethod used for task\n", "0.952021\tUNK UNK || local search\tREL\tmethod used for task\n", "0.952021\tlocal search || UNK\tREL\tmethod used for task\n", "0.952021\tUNK UNK || local search\tREL\tmethod used for task\n", "0.951947\tlinear UNK || genetic algorithm\tREL\tmethod used for task\n", "0.950822\tUNK UNK || UNK system\tREL\tmethod used for task\n", "0.950822\tUNK system || UNK\tREL\tmethod used for task\n", "0.950822\tUNK system || UNK\tREL\tmethod used for task\n", "0.950822\tUNK system || UNK\tREL\tmethod used for task\n", "0.950822\tUNK UNK || UNK system\tREL\tmethod used for task\n", "0.950822\tUNK UNK || UNK system\tREL\tmethod used for task\n", "0.950822\tUNK UNK || system\tREL\tmethod used for task\n", "0.950822\tUNK UNK || UNK system\tREL\tmethod used for task\n", "0.950822\tUNK UNK UNK || UNK system\tREL\tmethod used for task\n", "0.950822\tUNK UNK || UNK system\tREL\tmethod used for task\n", "0.950822\tUNK system || UNK\tREL\tmethod used for task\n", "0.950822\tUNK system || UNK\tREL\tmethod used for task\n", "0.950822\tUNK system || UNK\tREL\tmethod used for task\n", "0.950822\tUNK UNK UNK UNK system || UNK\tREL\tmethod used for task\n", "0.950822\tUNK UNK UNK || UNK system\tREL\tmethod used for task\n", "0.950822\tUNK UNK || UNK system\tREL\tmethod used for task\n", "0.950822\tUNK system || UNK\tREL\tmethod used for task\n", "0.950822\tUNK system || UNK\tREL\tmethod used for task\n", "0.950822\tUNK UNK system || UNK\tREL\tmethod used for task\n", "0.950822\tUNK system || UNK\tREL\tmethod used for task\n", "0.950822\tUNK UNK UNK system || UNK\tREL\tmethod used for task\n", "0.950822\tUNK UNK || UNK system\tREL\tmethod used for task\n", "0.950822\tUNK UNK || UNK UNK UNK system\tREL\tmethod used for task\n", "0.950822\tUNK UNK system || UNK\tREL\tmethod used for task\n", "0.950822\tUNK system || UNK\tREL\tmethod used for task\n", "0.950822\tUNK UNK UNK system || UNK\tREL\tmethod used for task\n", "0.950822\tUNK UNK UNK system || UNK\tREL\tmethod used for task\n", "0.950822\tUNK system || UNK\tREL\tmethod used for task\n", "0.950822\tUNK system || UNK\tREL\tmethod used for task\n", "0.950822\tUNK UNK || UNK system\tREL\tmethod used for task\n", "0.950822\tUNK UNK || UNK system\tREL\tmethod used for task\n", "0.950822\tUNK UNK || UNK system\tREL\tmethod used for task\n", "0.950822\tUNK UNK || UNK system\tREL\tmethod used for task\n", "0.950822\tUNK system || UNK\tREL\tmethod used for task\n", "0.950822\tUNK UNK system || UNK\tREL\tmethod used for task\n", "0.950822\tUNK system || UNK\tREL\tmethod used for task\n", "0.950822\tUNK UNK || UNK UNK UNK system\tREL\tmethod used for task\n", "0.950822\tUNK UNK || UNK system\tREL\tmethod used for task\n", "0.950822\tUNK UNK || UNK UNK system\tREL\tmethod used for task\n", "0.950822\tUNK UNK system || UNK\tREL\tmethod used for task\n", "0.950822\tUNK UNK || UNK system\tREL\tmethod used for task\n", "0.950822\tUNK UNK || UNK UNK system\tREL\tmethod used for task\n", "0.950822\tUNK UNK || UNK UNK system\tREL\tmethod used for task\n", "0.950822\tUNK system || UNK\tREL\tmethod used for task\n", "0.950822\tUNK system || UNK\tREL\tmethod used for task\n", "0.950822\tUNK system || UNK\tREL\tmethod used for task\n", "0.950822\tUNK system || UNK\tREL\tmethod used for task\n", "0.950822\tUNK UNK || UNK system\tREL\tmethod used for task\n", "0.950822\tUNK system || UNK\tREL\tmethod used for task\n", "0.950822\tUNK system || UNK\tREL\tmethod used for task\n", "0.950822\tUNK system || UNK\tREL\tmethod used for task\n", "0.950822\tUNK UNK || UNK system\tREL\tmethod used for task\n", "0.950822\tUNK system || UNK\tREL\tmethod used for task\n", "0.950822\tUNK UNK || UNK system\tREL\tmethod used for task\n", "0.950822\tUNK UNK system || UNK\tREL\tmethod used for task\n", "0.950822\tUNK UNK system || UNK\tREL\tmethod used for task\n", "0.950822\tUNK UNK system || UNK\tREL\tmethod used for task\n", "0.950822\tUNK UNK system || UNK\tREL\tmethod used for task\n", "0.950822\tUNK system || UNK\tREL\tmethod used for task\n", "0.950822\tUNK UNK || UNK system\tREL\tmethod used for task\n", "0.950822\tUNK UNK || UNK system\tREL\tmethod used for task\n", "0.950822\tUNK system || UNK\tREL\tmethod used for task\n", "0.950822\tUNK UNK || UNK system\tREL\tmethod used for task\n", "0.950822\tUNK UNK || UNK system\tREL\tmethod used for task\n", "0.950822\tUNK system || UNK\tREL\tmethod used for task\n", "0.950822\tUNK UNK system || UNK\tREL\tmethod used for task\n", "0.950822\tUNK UNK || UNK system\tREL\tmethod used for task\n", "0.950822\tUNK UNK || system\tREL\tmethod used for task\n", "0.950822\tUNK UNK || system\tREL\tmethod used for task\n", "0.950822\tsystem UNK || UNK\tREL\tmethod used for task\n", "0.950822\tUNK UNK || system\tREL\tmethod used for task\n", "0.950822\tUNK UNK || UNK system\tREL\tmethod used for task\n", "0.950822\tUNK UNK || UNK UNK system\tREL\tmethod used for task\n", "0.950822\tUNK system || UNK\tREL\tmethod used for task\n", "0.950822\tUNK system || UNK\tREL\tmethod used for task\n", "0.950822\tUNK UNK || UNK UNK system\tREL\tmethod used for task\n", "0.950822\tUNK system || UNK\tREL\tmethod used for task\n", "0.950822\tUNK UNK || system\tREL\tmethod used for task\n", "0.950822\tUNK system || UNK\tREL\tmethod used for task\n", "0.950822\tUNK system || UNK\tREL\tmethod used for task\n", "0.950822\tUNK UNK UNK || UNK system\tREL\tmethod used for task\n", "0.950822\tUNK system || UNK\tREL\tmethod used for task\n", "0.950521\tdomain UNK || relationship\tREL\tmethod used for task\n", "0.950238\tUNK performance || ensemble\tREL\tmethod used for task\n", "0.948892\tUNK UNK || UNK swarm algorithm\tREL\tmethod used for task\n", "0.948507\tgenetic algorithm || UNK parameters\tREL\tmethod used for task\n", "0.948437\tUNK models || model\tREL\tmethod used for task\n", "0.948437\tmodel UNK || UNK models\tREL\tmethod used for task\n", "0.948437\tmodel UNK || UNK models\tREL\tmethod used for task\n", "0.948437\tUNK model || UNK UNK models\tREL\tmethod used for task\n", "0.948437\tUNK models || UNK model\tREL\tmethod used for task\n", "0.948437\tUNK UNK models || UNK model\tREL\tmethod used for task\n", "0.948126\toptimal UNK || optimal\tREL\tmethod used for task\n", "0.948126\toptimal UNK || optimal\tREL\tmethod used for task\n", "0.948126\tUNK optimal UNK || optimal\tREL\tmethod used for task\n", "0.948126\toptimal UNK || optimal\tREL\tmethod used for task\n", "0.947490\tUNK model || model\tREL\tmethod used for task\n", "0.947490\tmodel UNK || UNK model\tREL\tmethod used for task\n", "0.947490\tUNK model || UNK model\tREL\tmethod used for task\n", "0.947490\tUNK model || UNK model\tREL\tmethod used for task\n", "0.947490\tUNK model || UNK model\tREL\tmethod used for task\n", "0.947490\tUNK model || UNK model\tREL\tmethod used for task\n", "0.947490\tUNK model || UNK model\tREL\tmethod used for task\n", "0.947428\tUNK model || UNK UNK UNK classifier\tREL\tmethod used for task\n", "0.947365\tUNK classifier || UNK classifier\tREL\tmethod used for task\n", "0.947362\tUNK models || supply\tREL\tmethod used for task\n", "0.946618\tUNK UNK method || heat\tREL\tmethod used for task\n", "0.946396\tUNK model || supply\tREL\tmethod used for task\n", "0.946088\tUNK algorithm || UNK design\tREL\tmethod used for task\n", "0.946088\tUNK design || UNK algorithm\tREL\tmethod used for task\n", "0.946088\tUNK algorithm || UNK design\tREL\tmethod used for task\n", "0.945292\tgenetic algorithm || UNK analysis\tREL\tmethod used for task\n", "0.945292\tgenetic algorithm || UNK analysis\tREL\tmethod used for task\n", "0.945292\tgenetic algorithm || UNK UNK analysis\tREL\tmethod used for task\n", "0.945194\tUNK models || UNK UNK method\tREL\tmethod used for task\n", "0.945095\tnonlinear models || UNK\tREL\tmethod used for task\n", "0.945016\tnonlinear UNK || pso\tREL\tmethod used for task\n", "0.944382\tUNK UNK techniques || UNK model\tREL\tmethod used for task\n", "0.944191\tUNK model || UNK method\tREL\tmethod used for task\n", "0.944191\tUNK method || UNK UNK UNK model\tREL\tmethod used for task\n", "0.943692\tgenetic algorithm || UNK simulation\tREL\tmethod used for task\n", "0.943692\tUNK simulation || genetic algorithm\tREL\tmethod used for task\n", "0.942423\tUNK model || UNK modeling\tREL\tmethod used for task\n", "0.941951\tUNK UNK || wireless sensor networks\tREL\tmethod used for task\n", "0.941951\twireless sensor networks || UNK\tREL\tmethod used for task\n", "0.941951\twireless sensor networks || UNK\tREL\tmethod used for task\n", "0.941951\tUNK UNK || wireless sensor networks\tREL\tmethod used for task\n", "0.941951\twireless sensor networks || UNK\tREL\tmethod used for task\n", "0.940698\tUNK UNK method || UNK method\tREL\tmethod used for task\n", "0.940698\tUNK UNK method || UNK method\tREL\tmethod used for task\n", "0.940698\tUNK method || UNK method\tREL\tmethod used for task\n", "0.940598\tpso algorithm || optimal solution\tREL\tmethod used for task\n", "0.940145\tshape models || UNK UNK model\tREL\tmethod used for task\n", "0.939927\tdata UNK || wireless sensor networks\tREL\tmethod used for task\n", "0.939860\toptimization problem || UNK variables\tREL\tmethod used for task\n", "0.939519\tUNK chain model || UNK\tREL\tmethod used for task\n", "0.938827\tUNK modeling method || UNK\tREL\tmethod used for task\n", "0.938827\tUNK modeling method || UNK\tREL\tmethod used for task\n", "0.938270\tsupply chain || UNK\tREL\tmethod used for task\n", "0.938270\tsupply chain || UNK\tREL\tmethod used for task\n", "0.938270\tsupply chain || UNK\tREL\tmethod used for task\n", "0.938270\tsupply chain || UNK\tREL\tmethod used for task\n", "0.938270\tsupply chain || UNK\tREL\tmethod used for task\n", "0.938270\tsupply chain || UNK\tREL\tmethod used for task\n", "0.938270\tUNK UNK || supply chain\tREL\tmethod used for task\n", "0.938270\tsupply chain || UNK\tREL\tmethod used for task\n", "0.938270\tsupply chain || UNK\tREL\tmethod used for task\n", "0.938270\tsupply chain || UNK\tREL\tmethod used for task\n", "0.938270\tsupply chain || UNK\tREL\tmethod used for task\n", "0.938270\tsupply chain || UNK\tREL\tmethod used for task\n", "0.938270\tUNK UNK || supply chain\tREL\tmethod used for task\n", "0.938270\tUNK UNK || supply chain\tREL\tmethod used for task\n", "0.938270\tsupply chain || UNK\tREL\tmethod used for task\n", "0.938270\tUNK UNK || supply chain\tREL\tmethod used for task\n", "0.938270\tsupply chain || UNK\tREL\tmethod used for task\n", "0.938270\tsupply chain || UNK\tREL\tmethod used for task\n", "0.938270\tUNK UNK || supply chain\tREL\tmethod used for task\n", "0.938270\tsupply chain || UNK\tREL\tmethod used for task\n", "0.938270\tsupply chain || UNK\tREL\tmethod used for task\n", "0.938270\tUNK UNK || supply chain\tREL\tmethod used for task\n", "0.938270\tUNK UNK || supply chain\tREL\tmethod used for task\n", "0.938270\tsupply chain || UNK\tREL\tmethod used for task\n", "0.938270\tsupply chain UNK || UNK\tREL\tmethod used for task\n", "0.938270\tsupply chain || UNK\tREL\tmethod used for task\n", "0.936900\tmodeling UNK || UNK modeling\tREL\tmethod used for task\n", "0.936900\tmodel UNK || UNK UNK theory\tREL\tmethod used for task\n", "0.936900\tUNK model UNK || UNK UNK theory\tREL\tmethod used for task\n", "0.936900\tUNK model UNK || UNK UNK theory\tREL\tmethod used for task\n", "0.935556\tUNK UNK UNK || neural networks\tREL\tmethod used for task\n", "0.935556\tUNK neural networks || UNK\tREL\tmethod used for task\n", "0.935556\tneural networks || UNK\tREL\tmethod used for task\n", "0.935556\tneural networks || UNK\tREL\tmethod used for task\n", "0.935556\tneural networks || UNK\tREL\tmethod used for task\n", "0.935556\tneural networks || UNK\tREL\tmethod used for task\n", "0.935556\tneural networks || UNK\tREL\tmethod used for task\n", "0.935556\tUNK UNK || neural networks\tREL\tmethod used for task\n", "0.935556\tUNK UNK || neural networks\tREL\tmethod used for task\n", "0.935556\tneural networks || UNK\tREL\tmethod used for task\n", "0.935556\tUNK UNK UNK || neural networks\tREL\tmethod used for task\n", "0.935556\tneural networks || UNK\tREL\tmethod used for task\n", "0.935556\tUNK UNK || UNK neural networks\tREL\tmethod used for task\n", "0.935357\tcase UNK || domain ontology\tREL\tmethod used for task\n", "0.935308\tUNK UNK || UNK optimization\tREL\tmethod used for task\n", "0.935308\tUNK UNK || optimization\tREL\tmethod used for task\n", "0.935308\tUNK UNK || optimization\tREL\tmethod used for task\n", "0.935308\toptimization UNK || UNK\tREL\tmethod used for task\n", "0.935308\tUNK UNK || UNK optimization\tREL\tmethod used for task\n", "0.935308\tUNK UNK || UNK optimization\tREL\tmethod used for task\n", "0.935308\tUNK UNK || optimization\tREL\tmethod used for task\n", "0.935308\tUNK UNK || UNK optimization\tREL\tmethod used for task\n", "0.935308\tUNK optimization || UNK\tREL\tmethod used for task\n", "0.935308\tUNK UNK UNK || UNK optimization\tREL\tmethod used for task\n", "0.935308\tUNK UNK || optimization\tREL\tmethod used for task\n", "0.935308\tUNK UNK || UNK optimization\tREL\tmethod used for task\n", "0.935308\toptimization UNK || UNK\tREL\tmethod used for task\n", "0.935308\tUNK UNK || UNK optimization\tREL\tmethod used for task\n", "0.935308\tUNK optimization || UNK\tREL\tmethod used for task\n", "0.935308\tUNK UNK || UNK optimization\tREL\tmethod used for task\n", "0.935308\tUNK optimization || UNK\tREL\tmethod used for task\n", "0.935308\tUNK UNK UNK || UNK UNK optimization\tREL\tmethod used for task\n", "0.935308\tUNK optimization || UNK\tREL\tmethod used for task\n", "0.935308\tUNK optimization || UNK\tREL\tmethod used for task\n", "0.935308\tUNK UNK || UNK UNK optimization\tREL\tmethod used for task\n", "0.935308\tUNK UNK || optimization\tREL\tmethod used for task\n", "0.935308\tUNK optimization || UNK\tREL\tmethod used for task\n", "0.935308\tUNK optimization || UNK\tREL\tmethod used for task\n", "0.935308\tUNK optimization || UNK\tREL\tmethod used for task\n", "0.935308\tUNK UNK || optimization\tREL\tmethod used for task\n", "0.935308\tUNK UNK || UNK optimization\tREL\tmethod used for task\n", "0.935308\tUNK optimization || UNK\tREL\tmethod used for task\n", "0.935308\tUNK optimization || UNK\tREL\tmethod used for task\n", "0.935308\tUNK UNK || UNK optimization\tREL\tmethod used for task\n", "0.935308\tUNK optimization || UNK\tREL\tmethod used for task\n", "0.935308\tUNK UNK || UNK optimization\tREL\tmethod used for task\n", "0.935308\tUNK optimization || UNK\tREL\tmethod used for task\n", "0.935308\toptimization UNK || UNK\tREL\tmethod used for task\n", "0.935292\tsystem performance || UNK condition\tREL\tmethod used for task\n", "0.935263\tshape model || UNK method\tREL\tmethod used for task\n", "0.933922\tUNK design || UNK models\tREL\tmethod used for task\n", "0.933467\tgenetic UNK || UNK UNK detection\tREL\tmethod used for task\n", "0.933130\toptimal design || UNK\tREL\tmethod used for task\n", "0.933069\toptimization UNK || data\tREL\tmethod used for task\n", "0.932822\tdesign optimization || UNK UNK analysis\tREL\tmethod used for task\n", "0.932728\tUNK model || UNK design\tREL\tmethod used for task\n", "0.932531\tmodel UNK || data mining technique\tREL\tmethod used for task\n", "0.931642\tlinear UNK UNK || UNK UNK set\tREL\tmethod used for task\n", "0.931091\tsystem parameters || UNK performance\tREL\tmethod used for task\n", "0.930903\tcost UNK || UNK optimization model\tREL\tmethod used for task\n", "0.929851\tdynamic models || UNK systems\tREL\tmethod used for task\n", "0.929851\tUNK models || dynamic systems\tREL\tmethod used for task\n", "0.929013\tdynamic systems || optimal\tREL\tmethod used for task\n", "0.928571\tUNK method || UNK design\tREL\tmethod used for task\n", "0.928571\tUNK design || UNK UNK method\tREL\tmethod used for task\n", "0.928165\tUNK UNK || intelligence method\tREL\tmethod used for task\n", "0.927168\tUNK optimization algorithm || UNK surface\tREL\tmethod used for task\n", "0.927058\tgenetic algorithm || neural network\tREL\tmethod used for task\n", "0.927058\tgenetic algorithm || UNK UNK UNK neural network\tREL\tmethod used for task\n", "0.926643\texperimental data || model\tREL\tmethod used for task\n", "0.926643\texperimental data || UNK model\tREL\tmethod used for task\n", "0.926643\tUNK model || experimental data\tREL\tmethod used for task\n", "0.926643\texperimental data || model\tREL\tmethod used for task\n", "0.926231\tensemble UNK || UNK\tREL\tmethod used for task\n", "0.926231\tUNK UNK || ensemble\tREL\tmethod used for task\n", "0.926231\tensemble UNK || UNK\tREL\tmethod used for task\n", "0.926231\tensemble UNK || UNK\tREL\tmethod used for task\n", "0.925986\tUNK UNK || detection problem\tREL\tmethod used for task\n", "0.925986\tUNK UNK || detection problem\tREL\tmethod used for task\n", "0.925526\tUNK UNK || UNK mining\tREL\tmethod used for task\n", "0.925526\tUNK UNK || UNK mining\tREL\tmethod used for task\n", "0.925526\tUNK mining || UNK\tREL\tmethod used for task\n", "0.925526\tUNK mining || UNK\tREL\tmethod used for task\n", "0.925378\tsimulated annealing || solution\tREL\tmethod used for task\n", "0.925378\tsimulated annealing || solution\tREL\tmethod used for task\n", "0.924717\texperimental UNK || UNK method\tREL\tmethod used for task\n", "0.924717\texperimental UNK || UNK method\tREL\tmethod used for task\n", "0.924512\tdynamic model || linear\tREL\tmethod used for task\n", "0.922977\tdata mining || UNK\tREL\tmethod used for task\n", "0.922977\tUNK UNK || data mining\tREL\tmethod used for task\n", "0.922977\tUNK data || UNK mining\tREL\tmethod used for task\n", "0.922977\tdata mining || UNK\tREL\tmethod used for task\n", "0.922977\tdata mining UNK || UNK\tREL\tmethod used for task\n", "0.922977\tUNK UNK || data mining\tREL\tmethod used for task\n", "0.922977\tUNK UNK || data mining\tREL\tmethod used for task\n", "0.922898\tartificial neural networks || UNK\tREL\tmethod used for task\n", "0.922898\tartificial neural networks || UNK\tREL\tmethod used for task\n", "0.922898\tUNK UNK || artificial neural networks\tREL\tmethod used for task\n", "0.922898\tartificial neural networks || UNK\tREL\tmethod used for task\n", "0.922898\tartificial neural networks || UNK\tREL\tmethod used for task\n", "0.922898\tartificial neural networks || UNK\tREL\tmethod used for task\n", "0.922898\tUNK UNK || artificial neural networks\tREL\tmethod used for task\n", "0.922898\tUNK UNK || artificial neural networks\tREL\tmethod used for task\n", "0.922898\tUNK UNK || artificial neural networks\tREL\tmethod used for task\n", "0.922898\tartificial neural networks || UNK\tREL\tmethod used for task\n", "0.922401\tUNK UNK techniques || experimental data\tREL\tmethod used for task\n", "0.922128\thybrid algorithm || numerical\tREL\tmethod used for task\n", "0.921347\tUNK clustering || UNK\tREL\tmethod used for task\n", "0.921347\tUNK clustering || UNK\tREL\tmethod used for task\n", "0.921347\tUNK UNK || UNK clustering\tREL\tmethod used for task\n", "0.921347\tUNK UNK clustering || UNK\tREL\tmethod used for task\n", "0.921347\tUNK clustering || UNK\tREL\tmethod used for task\n", "0.921347\tUNK clustering || UNK\tREL\tmethod used for task\n", "0.921347\tUNK clustering || UNK\tREL\tmethod used for task\n", "0.921347\tUNK clustering || UNK\tREL\tmethod used for task\n", "0.921347\tUNK clustering || UNK\tREL\tmethod used for task\n", "0.921347\tUNK clustering || UNK\tREL\tmethod used for task\n", "0.921347\tUNK clustering || UNK\tREL\tmethod used for task\n", "0.921347\tUNK UNK || clustering\tREL\tmethod used for task\n", "0.921347\tUNK clustering || UNK\tREL\tmethod used for task\n", "0.921347\tUNK UNK || UNK clustering\tREL\tmethod used for task\n", "0.921347\tclustering UNK || UNK\tREL\tmethod used for task\n", "0.921347\tUNK clustering || UNK\tREL\tmethod used for task\n", "0.921016\tUNK simulation || UNK search algorithm\tREL\tmethod used for task\n", "0.920451\tperformance UNK || hybrid\tREL\tmethod used for task\n", "0.920366\tUNK UNK || search performance\tREL\tmethod used for task\n", "0.920186\tneural network || ensemble classifier\tREL\tmethod used for task\n", "0.920035\tUNK algorithm || linear models\tREL\tmethod used for task\n", "0.919562\tUNK language UNK || UNK text\tREL\tmethod used for task\n", "0.919379\texperimental UNK || UNK shape models\tREL\tmethod used for task\n", "0.919147\tUNK domain || UNK performance\tREL\tmethod used for task\n", "0.918667\tdata clustering || UNK\tREL\tmethod used for task\n", "0.918667\tUNK clustering || UNK data\tREL\tmethod used for task\n", "0.918667\tclustering UNK || UNK data\tREL\tmethod used for task\n", "0.918667\tUNK clustering || UNK data\tREL\tmethod used for task\n", "0.918607\tUNK UNK UNK || case system\tREL\tmethod used for task\n", "0.918607\tUNK UNK || case system\tREL\tmethod used for task\n", "0.918607\tsystem UNK || case\tREL\tmethod used for task\n", "0.918131\tUNK algorithm || nonlinear systems\tREL\tmethod used for task\n", "0.917770\tdynamic UNK network || UNK clustering\tREL\tmethod used for task\n", "0.917559\tUNK UNK problem || UNK planning problem\tREL\tmethod used for task\n", "0.916516\tdynamic UNK || UNK detection\tREL\tmethod used for task\n", "0.916370\tgenetic algorithm || UNK space\tREL\tmethod used for task\n", "0.916349\tUNK UNK || UNK networks\tREL\tmethod used for task\n", "0.916349\tUNK UNK UNK networks || UNK\tREL\tmethod used for task\n", "0.916349\tUNK UNK UNK networks || UNK\tREL\tmethod used for task\n", "0.916349\tUNK UNK UNK networks || UNK\tREL\tmethod used for task\n", "0.916349\tUNK UNK || UNK networks\tREL\tmethod used for task\n", "0.916349\tUNK UNK || UNK networks\tREL\tmethod used for task\n", "0.916349\tUNK UNK || UNK networks\tREL\tmethod used for task\n", "0.916334\tUNK swarm optimization || UNK cost\tREL\tmethod used for task\n", "0.916288\treliability analysis || UNK system\tREL\tmethod used for task\n", "0.916121\tgenetic UNK || UNK\tREL\tmethod used for task\n", "0.916121\tgenetic UNK || UNK\tREL\tmethod used for task\n", "0.916121\tUNK UNK || genetic\tREL\tmethod used for task\n", "0.916121\tgenetic UNK || UNK\tREL\tmethod used for task\n", "0.916121\tgenetic UNK || UNK\tREL\tmethod used for task\n", "0.916121\tgenetic UNK || UNK\tREL\tmethod used for task\n", "0.916121\tgenetic UNK || UNK\tREL\tmethod used for task\n", "0.916121\tgenetic UNK || UNK\tREL\tmethod used for task\n", "0.915516\tUNK UNK || rough\tREL\tmethod used for task\n", "0.915516\tUNK UNK || rough\tREL\tmethod used for task\n", "0.915516\trough UNK || UNK\tREL\tmethod used for task\n", "0.915516\tUNK UNK || rough\tREL\tmethod used for task\n", "0.915516\tUNK UNK UNK || rough\tREL\tmethod used for task\n", "0.915516\tUNK UNK UNK || rough\tREL\tmethod used for task\n", "0.915516\trough UNK || UNK\tREL\tmethod used for task\n", "0.915516\trough UNK || UNK\tREL\tmethod used for task\n", "0.915296\tUNK swarm optimization || UNK UNK UNK neural network\tREL\tmethod used for task\n", "0.914782\tUNK UNK || strip\tREL\tmethod used for task\n", "0.914782\tUNK UNK || strip\tREL\tmethod used for task\n", "0.914190\tdesign UNK || design\tREL\tmethod used for task\n", "0.913515\tUNK networks || UNK data\tREL\tmethod used for task\n", "0.913515\tUNK networks || UNK data\tREL\tmethod used for task\n", "0.913515\tUNK networks || UNK data\tREL\tmethod used for task\n", "0.913114\tnumerical UNK || dynamic model\tREL\tmethod used for task\n", "0.913002\tUNK algorithm || model parameters\tREL\tmethod used for task\n", "0.910273\tUNK system || systems\tREL\tmethod used for task\n", "0.910273\tUNK UNK system || UNK UNK systems\tREL\tmethod used for task\n", "0.910273\tUNK system || UNK systems\tREL\tmethod used for task\n", "0.910051\tUNK algorithm || UNK detection\tREL\tmethod used for task\n", "0.910051\tUNK algorithm || UNK detection\tREL\tmethod used for task\n", "0.910051\tUNK UNK || UNK detection algorithm\tREL\tmethod used for task\n", "0.909220\tdynamic analysis || UNK method\tREL\tmethod used for task\n", "0.907820\toptimal solution || UNK optimization\tREL\tmethod used for task\n", "0.906885\tUNK UNK || UNK problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || UNK problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || UNK UNK problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || UNK problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || UNK problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || UNK UNK problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || UNK problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || UNK UNK problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || UNK problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || UNK UNK UNK problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || UNK UNK UNK problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || UNK UNK UNK problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || UNK UNK problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || UNK UNK problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || UNK UNK problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || UNK problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || UNK UNK problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || UNK problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || UNK UNK problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || UNK UNK problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || UNK UNK problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || UNK problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || UNK problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || UNK problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || UNK problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || UNK UNK problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || UNK problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || UNK problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || UNK problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || UNK UNK problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || UNK problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || UNK problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || UNK UNK UNK problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || UNK problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || UNK UNK problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || UNK problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || UNK problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || UNK UNK problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || UNK UNK problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || UNK problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || UNK problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || UNK UNK problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || UNK problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || UNK problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || UNK problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || UNK UNK problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || UNK problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || UNK problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || UNK problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK problem || UNK\tREL\tmethod used for task\n", "0.906885\tUNK problem || UNK\tREL\tmethod used for task\n", "0.906885\tUNK problem || UNK\tREL\tmethod used for task\n", "0.906885\tUNK UNK UNK UNK || UNK UNK problem\tREL\tmethod used for task\n", "0.906885\tUNK problem || UNK\tREL\tmethod used for task\n", "0.906885\tUNK problem || UNK\tREL\tmethod used for task\n", "0.906885\tUNK UNK problem || UNK\tREL\tmethod used for task\n", "0.906885\tUNK UNK problem || UNK\tREL\tmethod used for task\n", "0.906885\tUNK UNK problem || UNK\tREL\tmethod used for task\n", "0.906885\tUNK problem || UNK\tREL\tmethod used for task\n", "0.906885\tUNK UNK UNK || UNK problem\tREL\tmethod used for task\n", "0.906885\tUNK problem || UNK\tREL\tmethod used for task\n", "0.906885\tUNK UNK problem || UNK\tREL\tmethod used for task\n", "0.906885\tUNK UNK problem || UNK\tREL\tmethod used for task\n", "0.906885\tUNK problem || UNK\tREL\tmethod used for task\n", "0.906885\tproblem UNK || UNK\tREL\tmethod used for task\n", "0.906885\tUNK problem || UNK\tREL\tmethod used for task\n", "0.906885\tUNK UNK problem || UNK\tREL\tmethod used for task\n", "0.906885\tUNK UNK problem || UNK\tREL\tmethod used for task\n", "0.906885\tUNK UNK problem || UNK\tREL\tmethod used for task\n", "0.906885\tUNK problem || UNK\tREL\tmethod used for task\n", "0.906885\tproblem UNK UNK || UNK\tREL\tmethod used for task\n", "0.906885\tproblem UNK || UNK\tREL\tmethod used for task\n", "0.906885\tUNK problem || UNK\tREL\tmethod used for task\n", "0.906885\tUNK UNK UNK || UNK problem\tREL\tmethod used for task\n", "0.906885\tUNK problem || UNK\tREL\tmethod used for task\n", "0.906885\tUNK UNK || problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK problem || UNK\tREL\tmethod used for task\n", "0.906885\tproblem UNK UNK || UNK\tREL\tmethod used for task\n", "0.906885\tUNK UNK UNK || UNK problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK problem || UNK\tREL\tmethod used for task\n", "0.906885\tUNK UNK problem || UNK\tREL\tmethod used for task\n", "0.906885\tUNK UNK || problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK || problem\tREL\tmethod used for task\n", "0.906885\tUNK UNK problem || UNK\tREL\tmethod used for task\n", "0.906885\tUNK UNK problem || UNK\tREL\tmethod used for task\n", "0.906885\tUNK problem || UNK\tREL\tmethod used for task\n", "0.906885\tUNK UNK UNK || UNK UNK problem\tREL\tmethod used for task\n", "0.906087\tUNK system || UNK development\tREL\tmethod used for task\n", "0.906079\tUNK performance || UNK models\tREL\tmethod used for task\n", "0.906079\tperformance UNK || UNK models\tREL\tmethod used for task\n", "0.906079\tperformance UNK || UNK models\tREL\tmethod used for task\n", "0.906079\tUNK models || performance\tREL\tmethod used for task\n", "0.905257\tlinear system || UNK\tREL\tmethod used for task\n", "0.904496\tthe UNK || UNK\tREL\tmethod used for task\n", "0.904496\tthe UNK || UNK\tREL\tmethod used for task\n", "0.904433\tUNK UNK model || UNK performance\tREL\tmethod used for task\n", "0.904433\tmodel performance || UNK\tREL\tmethod used for task\n", "0.904433\tUNK model || UNK performance\tREL\tmethod used for task\n", "0.904433\tUNK UNK || model performance\tREL\tmethod used for task\n", "0.903163\tshape UNK || genetic\tREL\tmethod used for task\n", "0.900005\tUNK algorithm || planning problem\tREL\tmethod used for task\n", "0.899155\thybrid method || UNK analysis\tREL\tmethod used for task\n", "0.899046\tperformance UNK || UNK techniques\tREL\tmethod used for task\n", "0.899046\tperformance UNK || UNK techniques\tREL\tmethod used for task\n", "0.898829\tsystem parameters || UNK\tREL\tmethod used for task\n", "0.898718\tperformance UNK method || UNK\tREL\tmethod used for task\n", "0.898718\tUNK UNK || performance UNK method\tREL\tmethod used for task\n", "0.898718\tUNK performance || UNK method\tREL\tmethod used for task\n", "0.898718\tUNK method || UNK performance\tREL\tmethod used for task\n", "0.898602\tcost UNK || UNK set\tREL\tmethod used for task\n", "0.898510\thybrid algorithm || UNK UNK UNK neural network\tREL\tmethod used for task\n", "0.898483\tlanguage models || UNK\tREL\tmethod used for task\n", "0.898357\texperimental analysis || UNK UNK problem\tREL\tmethod used for task\n", "0.896138\tUNK UNK || surrogate\tREL\tmethod used for task\n", "0.895246\tdynamic UNK || UNK\tREL\tmethod used for task\n", "0.895246\tdynamic UNK || UNK\tREL\tmethod used for task\n", "0.895246\tUNK UNK || dynamic\tREL\tmethod used for task\n", "0.895246\tdynamic UNK || UNK\tREL\tmethod used for task\n", "0.895246\tdynamic UNK || UNK\tREL\tmethod used for task\n", "0.895246\tUNK UNK || dynamic\tREL\tmethod used for task\n", "0.895246\tdynamic UNK || UNK\tREL\tmethod used for task\n", "0.895246\tdynamic UNK || UNK\tREL\tmethod used for task\n", "0.895246\tdynamic UNK || UNK\tREL\tmethod used for task\n", "0.895246\tdynamic UNK || UNK\tREL\tmethod used for task\n", "0.895246\tdynamic UNK || UNK\tREL\tmethod used for task\n", "0.895170\tgenetic algorithm || linear UNK analysis\tREL\tmethod used for task\n", "0.892864\tUNK analysis || system\tREL\tmethod used for task\n", "0.892864\tUNK UNK analysis || UNK system\tREL\tmethod used for task\n", "0.891226\tUNK text || UNK\tREL\tmethod used for task\n", "0.891226\tUNK text || UNK\tREL\tmethod used for task\n", "0.891226\tUNK UNK || UNK text\tREL\tmethod used for task\n", "0.891226\tUNK text || UNK\tREL\tmethod used for task\n", "0.891226\tUNK text || UNK\tREL\tmethod used for task\n", "0.891226\tUNK UNK || text\tREL\tmethod used for task\n", "0.891019\tnumerical models || heat\tREL\tmethod used for task\n", "0.889426\tUNK chain model || computational\tREL\tmethod used for task\n", "0.888809\tUNK UNK detection || UNK model\tREL\tmethod used for task\n", "0.888584\tUNK systems || supply chain\tREL\tmethod used for task\n", "0.888438\tUNK cost || dynamic UNK model\tREL\tmethod used for task\n", "0.888288\tnumerical models || UNK models\tREL\tmethod used for task\n", "0.887999\tnetwork topology || UNK algorithm\tREL\tmethod used for task\n", "0.887496\tcomputational method || local\tREL\tmethod used for task\n", "0.887337\tUNK UNK || UNK algorithm\tREL\tmethod used for task\n", "0.887337\tUNK UNK || UNK algorithm\tREL\tmethod used for task\n", "0.887337\tUNK UNK || UNK algorithm\tREL\tmethod used for task\n", "0.887337\tUNK UNK || UNK algorithm\tREL\tmethod used for task\n", "0.887337\tUNK UNK || UNK algorithm\tREL\tmethod used for task\n", "0.887337\tUNK UNK || UNK algorithm\tREL\tmethod used for task\n", "0.887337\tUNK UNK || UNK algorithm\tREL\tmethod used for task\n", "0.887337\tUNK UNK || UNK algorithm\tREL\tmethod used for task\n", "0.887337\tUNK UNK || UNK algorithm\tREL\tmethod used for task\n", "0.887337\tUNK UNK || UNK algorithm\tREL\tmethod used for task\n", "0.887337\tUNK UNK || UNK algorithm\tREL\tmethod used for task\n", "0.887337\tUNK UNK || UNK algorithm\tREL\tmethod used for task\n", "0.887337\tUNK UNK || UNK algorithm\tREL\tmethod used for task\n", "0.887337\tUNK UNK || UNK algorithm\tREL\tmethod used for task\n", "0.887337\tUNK UNK || UNK UNK algorithm\tREL\tmethod used for task\n", "0.887337\tUNK UNK || UNK algorithm\tREL\tmethod used for task\n", "0.887337\tUNK UNK || UNK algorithm\tREL\tmethod used for task\n", "0.887337\tUNK UNK || UNK UNK algorithm\tREL\tmethod used for task\n", "0.887337\tUNK UNK || UNK algorithm\tREL\tmethod used for task\n", "0.887337\tUNK UNK || UNK UNK algorithm\tREL\tmethod used for task\n", "0.887337\tUNK UNK || UNK algorithm\tREL\tmethod used for task\n", "0.887337\tUNK UNK || UNK algorithm\tREL\tmethod used for task\n", "0.887337\tUNK UNK || UNK UNK algorithm\tREL\tmethod used for task\n", "0.887337\tUNK UNK || UNK algorithm\tREL\tmethod used for task\n", "0.887337\tUNK UNK || UNK algorithm\tREL\tmethod used for task\n", "0.887337\tUNK UNK || UNK algorithm\tREL\tmethod used for task\n", "0.887337\tUNK UNK || UNK algorithm\tREL\tmethod used for task\n", "0.887337\tUNK UNK || UNK UNK algorithm\tREL\tmethod used for task\n", "0.887337\tUNK UNK || UNK algorithm\tREL\tmethod used for task\n", "0.887337\tUNK UNK || UNK algorithm\tREL\tmethod used for task\n", "0.887337\tUNK UNK || UNK algorithm\tREL\tmethod used for task\n", "0.887337\tUNK UNK || UNK UNK algorithm\tREL\tmethod used for task\n", "0.887337\tUNK UNK || UNK UNK algorithm\tREL\tmethod used for task\n", "0.887337\tUNK UNK || UNK algorithm\tREL\tmethod used for task\n", "0.887337\tUNK UNK || UNK algorithm\tREL\tmethod used for task\n", "0.887337\tUNK UNK || UNK algorithm\tREL\tmethod used for task\n", "0.887337\tUNK UNK || UNK algorithm\tREL\tmethod used for task\n", "0.887337\tUNK UNK || UNK algorithm\tREL\tmethod used for task\n", "0.887337\tUNK UNK || UNK algorithm\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK UNK UNK || UNK algorithm\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK UNK UNK || UNK algorithm\tREL\tmethod used for task\n", "0.887337\tUNK UNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK UNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK UNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK UNK UNK || UNK algorithm\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.887337\tUNK algorithm || UNK\tREL\tmethod used for task\n", "0.886030\tmodel parameters || UNK UNK method\tREL\tmethod used for task\n", "0.885140\tUNK UNK UNK || building\tREL\tmethod used for task\n", "0.885140\tUNK UNK UNK || building\tREL\tmethod used for task\n", "0.885140\tbuilding UNK || UNK\tREL\tmethod used for task\n", "0.885140\tbuilding UNK || UNK\tREL\tmethod used for task\n", "0.885140\tbuilding UNK || UNK\tREL\tmethod used for task\n", "0.885140\tUNK UNK || building\tREL\tmethod used for task\n", "0.885140\tUNK UNK || building\tREL\tmethod used for task\n", "0.884044\tUNK algorithm || search space\tREL\tmethod used for task\n", "0.883830\thybrid UNK || UNK\tREL\tmethod used for task\n", "0.883830\thybrid UNK || UNK\tREL\tmethod used for task\n", "0.883830\thybrid UNK UNK || UNK\tREL\tmethod used for task\n", "0.883830\thybrid UNK || UNK\tREL\tmethod used for task\n", "0.883711\tUNK UNK || UNK UNK search\tREL\tmethod used for task\n", "0.883711\tUNK search || UNK\tREL\tmethod used for task\n", "0.883711\tUNK UNK || search\tREL\tmethod used for task\n", "0.883711\tUNK UNK || search\tREL\tmethod used for task\n", "0.883711\tUNK UNK || search\tREL\tmethod used for task\n", "0.883711\tUNK UNK || UNK search\tREL\tmethod used for task\n", "0.883711\tUNK UNK search || UNK\tREL\tmethod used for task\n", "0.883711\tUNK search || UNK\tREL\tmethod used for task\n", "0.883711\tUNK UNK || UNK search\tREL\tmethod used for task\n", "0.883711\tUNK search || UNK\tREL\tmethod used for task\n", "0.883711\tUNK search || UNK\tREL\tmethod used for task\n", "0.883711\tUNK UNK search || UNK\tREL\tmethod used for task\n", "0.883711\tUNK UNK search || UNK\tREL\tmethod used for task\n", "0.883711\tUNK search || UNK\tREL\tmethod used for task\n", "0.883711\tUNK UNK || UNK search\tREL\tmethod used for task\n", "0.883711\tUNK UNK || UNK search\tREL\tmethod used for task\n", "0.883711\tUNK UNK || search\tREL\tmethod used for task\n", "0.883711\tUNK search || UNK\tREL\tmethod used for task\n", "0.883711\tUNK UNK || search\tREL\tmethod used for task\n", "0.883711\tUNK search || UNK\tREL\tmethod used for task\n", "0.883711\tUNK UNK || UNK search\tREL\tmethod used for task\n", "0.883645\tUNK algorithm || data\tREL\tmethod used for task\n", "0.883645\tUNK algorithm || data\tREL\tmethod used for task\n", "0.883645\tdata UNK || UNK algorithm\tREL\tmethod used for task\n", "0.883645\tdata UNK || UNK algorithm\tREL\tmethod used for task\n", "0.883645\tUNK algorithm || data\tREL\tmethod used for task\n", "0.883645\tUNK algorithm || UNK data\tREL\tmethod used for task\n", "0.883645\tUNK algorithm || UNK data\tREL\tmethod used for task\n", "0.883645\tUNK algorithm || UNK data\tREL\tmethod used for task\n", "0.882881\tUNK set || UNK space\tREL\tmethod used for task\n", "0.882281\tUNK UNK UNK || UNK detection method\tREL\tmethod used for task\n", "0.882165\tcomputational UNK || optimization\tREL\tmethod used for task\n", "0.882083\tnonlinear UNK || UNK detection\tREL\tmethod used for task\n", "0.882003\tUNK domain || UNK\tREL\tmethod used for task\n", "0.882003\tUNK UNK || UNK domain\tREL\tmethod used for task\n", "0.882003\tUNK UNK || UNK domain\tREL\tmethod used for task\n", "0.882003\tUNK domain || UNK\tREL\tmethod used for task\n", "0.882003\tdomain UNK || UNK\tREL\tmethod used for task\n", "0.882003\tdomain UNK || UNK\tREL\tmethod used for task\n", "0.882003\tUNK UNK || UNK domain\tREL\tmethod used for task\n", "0.882003\tUNK UNK || UNK domain\tREL\tmethod used for task\n", "0.882003\tUNK UNK || UNK domain\tREL\tmethod used for task\n", "0.881957\tUNK UNK || optimisation\tREL\tmethod used for task\n", "0.881957\tUNK UNK || optimisation\tREL\tmethod used for task\n", "0.881488\tneural UNK system || UNK sensor\tREL\tmethod used for task\n", "0.880326\tUNK search || search space\tREL\tmethod used for task\n", "0.880038\thybrid UNK || UNK data\tREL\tmethod used for task\n", "0.879916\tdata UNK || UNK search\tREL\tmethod used for task\n", "0.879838\tUNK quality || UNK system design\tREL\tmethod used for task\n", "0.878066\tsystem UNK || product\tREL\tmethod used for task\n", "0.878017\tlocal UNK || detection\tREL\tmethod used for task\n", "0.877261\texperimental analysis || UNK algorithm\tREL\tmethod used for task\n", "0.877226\toptimization UNK || linear\tREL\tmethod used for task\n", "0.875542\tUNK network || system performance\tREL\tmethod used for task\n", "0.874922\tcase UNK || data mining\tREL\tmethod used for task\n", "0.873485\texperimental analysis || hybrid\tREL\tmethod used for task\n", "0.872952\texperimental UNK || UNK performance\tREL\tmethod used for task\n", "0.871078\tnetwork model || problem\tREL\tmethod used for task\n", "0.869171\tUNK UNK optimization || UNK parameters\tREL\tmethod used for task\n", "0.869171\toptimization UNK || UNK parameters\tREL\tmethod used for task\n", "0.869011\toptimal solution || UNK problem\tREL\tmethod used for task\n", "0.868345\tUNK UNK || quantum\tREL\tmethod used for task\n", "0.867060\tUNK UNK || heat\tREL\tmethod used for task\n", "0.867060\theat UNK || UNK\tREL\tmethod used for task\n", "0.867060\theat UNK || UNK\tREL\tmethod used for task\n", "0.867060\tUNK UNK || heat\tREL\tmethod used for task\n", "0.867060\tUNK UNK || heat\tREL\tmethod used for task\n", "0.864335\tUNK design || model parameters\tREL\tmethod used for task\n", "0.863821\tUNK models || UNK\tREL\tmethod used for task\n", "0.863821\tUNK UNK || UNK models\tREL\tmethod used for task\n", "0.863821\tUNK UNK || UNK models\tREL\tmethod used for task\n", "0.863821\tUNK UNK || UNK UNK models\tREL\tmethod used for task\n", "0.863821\tUNK UNK || UNK models\tREL\tmethod used for task\n", "0.863821\tUNK UNK || UNK models\tREL\tmethod used for task\n", "0.863821\tUNK models || UNK\tREL\tmethod used for task\n", "0.863821\tUNK models || UNK\tREL\tmethod used for task\n", "0.863821\tUNK models || UNK\tREL\tmethod used for task\n", "0.863821\tUNK models || UNK\tREL\tmethod used for task\n", "0.863821\tUNK models || UNK\tREL\tmethod used for task\n", "0.863821\tUNK UNK models || UNK\tREL\tmethod used for task\n", "0.863821\tUNK UNK || UNK UNK models\tREL\tmethod used for task\n", "0.863821\tUNK UNK || UNK models\tREL\tmethod used for task\n", "0.863821\tUNK UNK UNK || UNK models\tREL\tmethod used for task\n", "0.863821\tUNK UNK || UNK models\tREL\tmethod used for task\n", "0.863821\tUNK models || UNK\tREL\tmethod used for task\n", "0.863821\tUNK UNK || UNK models\tREL\tmethod used for task\n", "0.863821\tUNK UNK || UNK models\tREL\tmethod used for task\n", "0.863821\tUNK UNK || UNK UNK models\tREL\tmethod used for task\n", "0.863821\tUNK UNK || UNK models\tREL\tmethod used for task\n", "0.863821\tUNK UNK || UNK models\tREL\tmethod used for task\n", "0.863821\tUNK UNK || UNK models\tREL\tmethod used for task\n", "0.863821\tUNK models || UNK\tREL\tmethod used for task\n", "0.863821\tUNK UNK models || UNK\tREL\tmethod used for task\n", "0.863821\tUNK UNK || UNK models\tREL\tmethod used for task\n", "0.863821\tUNK models || UNK\tREL\tmethod used for task\n", "0.863821\tUNK models || UNK\tREL\tmethod used for task\n", "0.863821\tUNK UNK || UNK models\tREL\tmethod used for task\n", "0.863821\tUNK models || UNK\tREL\tmethod used for task\n", "0.863821\tUNK UNK || UNK models\tREL\tmethod used for task\n", "0.863821\tUNK UNK || UNK models\tREL\tmethod used for task\n", "0.863821\tUNK UNK || UNK models\tREL\tmethod used for task\n", "0.863821\tUNK models || UNK\tREL\tmethod used for task\n", "0.863821\tUNK UNK || UNK models\tREL\tmethod used for task\n", "0.863821\tUNK UNK || UNK models\tREL\tmethod used for task\n", "0.863821\tUNK models || UNK\tREL\tmethod used for task\n", "0.863821\tUNK UNK || UNK models\tREL\tmethod used for task\n", "0.863821\tUNK UNK || UNK models\tREL\tmethod used for task\n", "0.863821\tUNK models || UNK\tREL\tmethod used for task\n", "0.863821\tUNK UNK || UNK UNK models\tREL\tmethod used for task\n", "0.863821\tUNK models || UNK\tREL\tmethod used for task\n", "0.863821\tUNK models || UNK\tREL\tmethod used for task\n", "0.863821\tUNK models || UNK\tREL\tmethod used for task\n", "0.863821\tUNK models || UNK\tREL\tmethod used for task\n", "0.863821\tUNK UNK models || UNK\tREL\tmethod used for task\n", "0.863821\tUNK UNK || UNK models\tREL\tmethod used for task\n", "0.863821\tUNK UNK || UNK models\tREL\tmethod used for task\n", "0.863810\tUNK UNK regression || UNK genetic algorithm\tREL\tmethod used for task\n", "0.862311\tUNK optimal UNK || UNK\tREL\tmethod used for task\n", "0.862311\toptimal UNK || UNK\tREL\tmethod used for task\n", "0.862311\toptimal UNK || UNK\tREL\tmethod used for task\n", "0.862311\tUNK UNK || optimal\tREL\tmethod used for task\n", "0.862311\tUNK UNK || optimal\tREL\tmethod used for task\n", "0.862311\toptimal UNK || UNK\tREL\tmethod used for task\n", "0.862311\toptimal UNK || UNK\tREL\tmethod used for task\n", "0.862311\toptimal UNK || UNK\tREL\tmethod used for task\n", "0.862311\toptimal UNK || UNK\tREL\tmethod used for task\n", "0.862311\toptimal UNK || UNK\tREL\tmethod used for task\n", "0.862311\tUNK UNK || optimal\tREL\tmethod used for task\n", "0.862311\toptimal UNK || UNK\tREL\tmethod used for task\n", "0.862311\tUNK UNK || optimal\tREL\tmethod used for task\n", "0.862311\toptimal UNK || UNK\tREL\tmethod used for task\n", "0.862311\toptimal UNK || UNK\tREL\tmethod used for task\n", "0.862311\toptimal UNK || UNK\tREL\tmethod used for task\n", "0.862311\tUNK UNK || UNK optimal\tREL\tmethod used for task\n", "0.862311\toptimal UNK || UNK\tREL\tmethod used for task\n", "0.862311\tUNK optimal UNK || UNK\tREL\tmethod used for task\n", "0.862311\toptimal UNK || UNK\tREL\tmethod used for task\n", "0.862311\toptimal UNK || UNK\tREL\tmethod used for task\n", "0.862311\tUNK UNK || optimal\tREL\tmethod used for task\n", "0.862311\toptimal UNK || UNK\tREL\tmethod used for task\n", "0.862311\tUNK UNK || optimal\tREL\tmethod used for task\n", "0.862311\toptimal UNK || UNK\tREL\tmethod used for task\n", "0.861726\toptimization UNK || UNK analysis\tREL\tmethod used for task\n", "0.861547\tUNK UNK || UNK model\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK UNK || UNK model\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK UNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK UNK || UNK model\tREL\tmethod used for task\n", "0.861547\tUNK UNK || model\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tmodel UNK || UNK\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK UNK || model\tREL\tmethod used for task\n", "0.861547\tUNK UNK || model\tREL\tmethod used for task\n", "0.861547\tUNK UNK || UNK model\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tmodel UNK || UNK\tREL\tmethod used for task\n", "0.861547\tmodel UNK || UNK\tREL\tmethod used for task\n", "0.861547\tmodel UNK || UNK\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK UNK || UNK model\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK UNK || UNK model\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tmodel UNK || UNK\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK UNK || model\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK UNK || UNK model\tREL\tmethod used for task\n", "0.861547\tUNK UNK model || UNK\tREL\tmethod used for task\n", "0.861547\tmodel UNK || UNK\tREL\tmethod used for task\n", "0.861547\tmodel UNK || UNK\tREL\tmethod used for task\n", "0.861547\tUNK UNK || UNK model\tREL\tmethod used for task\n", "0.861547\tmodel UNK || UNK\tREL\tmethod used for task\n", "0.861547\tmodel UNK || UNK\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK UNK || UNK model\tREL\tmethod used for task\n", "0.861547\tUNK UNK || model\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK UNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK UNK || UNK UNK UNK model\tREL\tmethod used for task\n", "0.861547\tUNK UNK || UNK model\tREL\tmethod used for task\n", "0.861547\tUNK UNK || UNK model\tREL\tmethod used for task\n", "0.861547\tUNK UNK || UNK model\tREL\tmethod used for task\n", "0.861547\tUNK UNK || UNK model\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK UNK || model\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK UNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK UNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK UNK || UNK model\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tmodel UNK || UNK\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK UNK || UNK model\tREL\tmethod used for task\n", "0.861547\tUNK UNK || UNK model\tREL\tmethod used for task\n", "0.861547\tUNK UNK || UNK model\tREL\tmethod used for task\n", "0.861547\tUNK UNK || UNK model\tREL\tmethod used for task\n", "0.861547\tUNK UNK || UNK UNK model\tREL\tmethod used for task\n", "0.861547\tUNK UNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK UNK || UNK model\tREL\tmethod used for task\n", "0.861547\tUNK UNK || UNK UNK model\tREL\tmethod used for task\n", "0.861547\tUNK UNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK UNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK UNK || UNK model\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK UNK || model\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tmodel UNK || UNK\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK UNK || UNK model\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK UNK || UNK model\tREL\tmethod used for task\n", "0.861547\tUNK UNK UNK || model\tREL\tmethod used for task\n", "0.861547\tmodel UNK || UNK\tREL\tmethod used for task\n", "0.861547\tUNK UNK || model\tREL\tmethod used for task\n", "0.861547\tUNK UNK || UNK model\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK UNK || UNK model\tREL\tmethod used for task\n", "0.861547\tUNK UNK || UNK UNK model\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK UNK || UNK model\tREL\tmethod used for task\n", "0.861547\tUNK UNK || UNK model\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK UNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK UNK || UNK model\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK UNK || UNK model\tREL\tmethod used for task\n", "0.861547\tUNK UNK || UNK model\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK UNK || UNK model\tREL\tmethod used for task\n", "0.861547\tUNK UNK || model\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK UNK || UNK model\tREL\tmethod used for task\n", "0.861547\tUNK UNK || model\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK UNK || UNK model\tREL\tmethod used for task\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK UNK || UNK model\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK UNK || UNK model\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK UNK || UNK UNK model\tREL\tmethod used for task\n", "0.861547\tUNK UNK || UNK model\tREL\tmethod used for task\n", "0.861547\tmodel UNK || UNK\tREL\tmethod used for task\n", "0.861547\tmodel UNK || UNK\tREL\tmethod used for task\n", "0.861547\tUNK UNK || UNK model\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK UNK || UNK model\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tmodel UNK || UNK\tREL\tmethod used for task\n", "0.861547\tUNK UNK || UNK UNK model\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK UNK || UNK model\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK UNK || model\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK UNK || UNK model\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK UNK || UNK model\tREL\tmethod used for task\n", "0.861547\tUNK UNK || model\tREL\tmethod used for task\n", "0.861547\tUNK UNK || model\tREL\tmethod used for task\n", "0.861547\tUNK UNK model || UNK\tREL\tmethod used for task\n", "0.861547\tUNK UNK || UNK model\tREL\tmethod used for task\n", "0.861397\tUNK UNK || UNK classifier\tREL\tmethod used for task\n", "0.861397\tUNK UNK || UNK classifier\tREL\tmethod used for task\n", "0.861397\tUNK classifier || UNK\tREL\tmethod used for task\n", "0.861397\tUNK classifier || UNK\tREL\tmethod used for task\n", "0.861397\tUNK UNK || UNK classifier\tREL\tmethod used for task\n", "0.861397\tUNK UNK UNK || UNK classifier\tREL\tmethod used for task\n", "0.860961\tdata structures || surface\tREL\tmethod used for task\n", "0.860072\tUNK systems || UNK clustering\tREL\tmethod used for task\n", "0.859747\tUNK system || neural network\tREL\tmethod used for task\n", "0.859480\tUNK data || UNK models\tREL\tmethod used for task\n", "0.859480\tUNK models || UNK data\tREL\tmethod used for task\n", "0.859480\tUNK data || UNK models\tREL\tmethod used for task\n", "0.859480\tdata UNK || UNK models\tREL\tmethod used for task\n", "0.859295\tUNK search || UNK construction\tREL\tmethod used for task\n", "0.858928\tUNK UNK || supply\tREL\tmethod used for task\n", "0.858928\tUNK UNK || supply\tREL\tmethod used for task\n", "0.858928\tUNK UNK || supply\tREL\tmethod used for task\n", "0.858928\tUNK UNK || supply\tREL\tmethod used for task\n", "0.858048\tUNK simulation || UNK optimization\tREL\tmethod used for task\n", "0.857481\tdynamic models || UNK variables\tREL\tmethod used for task\n", "0.857145\tUNK data || UNK model\tREL\tmethod used for task\n", "0.857145\tUNK data || UNK model\tREL\tmethod used for task\n", "0.857145\tUNK model || UNK data\tREL\tmethod used for task\n", "0.857145\tdata model || UNK\tREL\tmethod used for task\n", "0.857145\tdata model || UNK\tREL\tmethod used for task\n", "0.857145\tdata UNK || model\tREL\tmethod used for task\n", "0.855119\tdynamic model || UNK variables\tREL\tmethod used for task\n", "0.854994\tUNK solution || dynamic models\tREL\tmethod used for task\n", "0.854133\tUNK UNK || UNK UNK techniques\tREL\tmethod used for task\n", "0.854133\tUNK UNK || UNK UNK techniques\tREL\tmethod used for task\n", "0.854133\tUNK UNK techniques || UNK\tREL\tmethod used for task\n", "0.854133\tUNK UNK UNK || UNK techniques\tREL\tmethod used for task\n", "0.854133\tUNK UNK || UNK UNK techniques\tREL\tmethod used for task\n", "0.854133\tUNK UNK || UNK UNK techniques\tREL\tmethod used for task\n", "0.854133\tUNK UNK UNK || UNK techniques\tREL\tmethod used for task\n", "0.854133\tUNK UNK techniques || UNK\tREL\tmethod used for task\n", "0.854133\tUNK UNK || UNK UNK techniques\tREL\tmethod used for task\n", "0.854133\tUNK UNK || UNK UNK techniques\tREL\tmethod used for task\n", "0.854133\tUNK UNK techniques || UNK\tREL\tmethod used for task\n", "0.854133\tUNK UNK || UNK techniques\tREL\tmethod used for task\n", "0.854133\tUNK UNK || UNK techniques\tREL\tmethod used for task\n", "0.854133\tUNK UNK || UNK techniques\tREL\tmethod used for task\n", "0.854133\tUNK UNK || UNK techniques\tREL\tmethod used for task\n", "0.854133\tUNK UNK || UNK techniques\tREL\tmethod used for task\n", "0.853682\tUNK UNK UNK || UNK UNK method\tREL\tmethod used for task\n", "0.853682\tUNK UNK UNK || UNK UNK method\tREL\tmethod used for task\n", "0.853682\tUNK UNK || UNK UNK method\tREL\tmethod used for task\n", "0.853682\tUNK UNK method || UNK\tREL\tmethod used for task\n", "0.853682\tUNK UNK || UNK method\tREL\tmethod used for task\n", "0.853682\tUNK UNK || UNK method\tREL\tmethod used for task\n", "0.853682\tUNK UNK || UNK method\tREL\tmethod used for task\n", "0.853682\tUNK method || UNK\tREL\tmethod used for task\n", "0.853682\tUNK method || UNK\tREL\tmethod used for task\n", "0.853682\tUNK UNK || UNK method\tREL\tmethod used for task\n", "0.853682\tUNK UNK || UNK method\tREL\tmethod used for task\n", "0.853682\tUNK method || UNK\tREL\tmethod used for task\n", "0.853682\tUNK UNK || UNK method\tREL\tmethod used for task\n", "0.853682\tUNK UNK || UNK method\tREL\tmethod used for task\n", "0.853682\tUNK UNK || UNK method\tREL\tmethod used for task\n", "0.853682\tUNK method || UNK\tREL\tmethod used for task\n", "0.853682\tUNK UNK || UNK UNK method\tREL\tmethod used for task\n", "0.853682\tUNK UNK || UNK UNK method\tREL\tmethod used for task\n", "0.853682\tUNK UNK || UNK method\tREL\tmethod used for task\n", "0.853682\tUNK UNK || UNK UNK method\tREL\tmethod used for task\n", "0.853682\tUNK UNK UNK || UNK method\tREL\tmethod used for task\n", "0.853682\tUNK method || UNK\tREL\tmethod used for task\n", "0.853682\tUNK method || UNK\tREL\tmethod used for task\n", "0.853682\tUNK UNK || UNK UNK method\tREL\tmethod used for task\n", "0.853682\tUNK UNK || UNK UNK method\tREL\tmethod used for task\n", "0.853682\tUNK UNK || UNK UNK method\tREL\tmethod used for task\n", "0.853682\tUNK UNK || UNK UNK method\tREL\tmethod used for task\n", "0.853682\tUNK UNK method || UNK\tREL\tmethod used for task\n", "0.853682\tUNK UNK method || UNK\tREL\tmethod used for task\n", "0.853682\tUNK UNK UNK || UNK method\tREL\tmethod used for task\n", "0.853682\tUNK method || UNK\tREL\tmethod used for task\n", "0.853682\tUNK UNK || UNK method\tREL\tmethod used for task\n", "0.853682\tUNK method || UNK\tREL\tmethod used for task\n", "0.853682\tUNK UNK || UNK UNK method\tREL\tmethod used for task\n", "0.853682\tUNK UNK || UNK UNK method\tREL\tmethod used for task\n", "0.853682\tUNK UNK || UNK method\tREL\tmethod used for task\n", "0.853682\tUNK UNK || UNK method\tREL\tmethod used for task\n", "0.853682\tUNK UNK || UNK UNK method\tREL\tmethod used for task\n", "0.853682\tUNK UNK || UNK method\tREL\tmethod used for task\n", "0.853682\tUNK UNK || UNK method\tREL\tmethod used for task\n", "0.853682\tUNK UNK || UNK UNK method\tREL\tmethod used for task\n", "0.853682\tUNK method || UNK\tREL\tmethod used for task\n", "0.853682\tUNK UNK || UNK method\tREL\tmethod used for task\n", "0.853682\tUNK UNK || UNK UNK method\tREL\tmethod used for task\n", "0.853682\tUNK UNK UNK || UNK method\tREL\tmethod used for task\n", "0.853682\tUNK method || UNK\tREL\tmethod used for task\n", "0.853682\tUNK method || UNK\tREL\tmethod used for task\n", "0.853682\tUNK UNK || UNK method\tREL\tmethod used for task\n", "0.853682\tUNK UNK || UNK method\tREL\tmethod used for task\n", "0.853682\tUNK UNK || UNK method\tREL\tmethod used for task\n", "0.853682\tUNK UNK || UNK UNK method\tREL\tmethod used for task\n", "0.853682\tUNK UNK || UNK method\tREL\tmethod used for task\n", "0.853682\tUNK UNK || UNK method\tREL\tmethod used for task\n", "0.853682\tUNK UNK || UNK method\tREL\tmethod used for task\n", "0.853682\tUNK method || UNK\tREL\tmethod used for task\n", "0.853682\tUNK method || UNK\tREL\tmethod used for task\n", "0.853682\tUNK UNK || UNK method\tREL\tmethod used for task\n", "0.853682\tUNK UNK || UNK method\tREL\tmethod used for task\n", "0.853682\tUNK UNK || UNK method\tREL\tmethod used for task\n", "0.853682\tUNK UNK || UNK method\tREL\tmethod used for task\n", "0.853682\tUNK UNK method || UNK\tREL\tmethod used for task\n", "0.853682\tUNK UNK || UNK method\tREL\tmethod used for task\n", "0.853682\tUNK UNK || UNK method\tREL\tmethod used for task\n", "0.853682\tUNK UNK || UNK method\tREL\tmethod used for task\n", "0.853682\tUNK UNK || UNK UNK method\tREL\tmethod used for task\n", "0.853682\tUNK UNK || UNK method\tREL\tmethod used for task\n", "0.853682\tUNK method || UNK\tREL\tmethod used for task\n", "0.853682\tUNK UNK method || UNK\tREL\tmethod used for task\n", "0.853682\tUNK UNK || UNK method\tREL\tmethod used for task\n", "0.853682\tUNK method || UNK\tREL\tmethod used for task\n", "0.853682\tUNK UNK || UNK method\tREL\tmethod used for task\n", "0.853682\tUNK UNK || UNK method\tREL\tmethod used for task\n", "0.853682\tUNK method || UNK\tREL\tmethod used for task\n", "0.853682\tUNK method || UNK\tREL\tmethod used for task\n", "0.853682\tUNK method || UNK\tREL\tmethod used for task\n", "0.853682\tUNK method || UNK\tREL\tmethod used for task\n", "0.853682\tUNK method || UNK\tREL\tmethod used for task\n", "0.853682\tUNK UNK || UNK method\tREL\tmethod used for task\n", "0.853682\tUNK method || UNK\tREL\tmethod used for task\n", "0.853682\tUNK UNK || UNK method\tREL\tmethod used for task\n", "0.853682\tUNK UNK || UNK method\tREL\tmethod used for task\n", "0.853682\tUNK method || UNK\tREL\tmethod used for task\n", "0.853682\tUNK method || UNK\tREL\tmethod used for task\n", "0.853682\tUNK UNK method || UNK\tREL\tmethod used for task\n", "0.853682\tUNK method || UNK\tREL\tmethod used for task\n", "0.853682\tUNK UNK || UNK method\tREL\tmethod used for task\n", "0.853682\tUNK method || UNK\tREL\tmethod used for task\n", "0.853682\tUNK method || UNK\tREL\tmethod used for task\n", "0.853682\tUNK method || UNK\tREL\tmethod used for task\n", "0.853682\tUNK method || UNK\tREL\tmethod used for task\n", "0.853682\tUNK UNK || UNK method\tREL\tmethod used for task\n", "0.853682\tUNK UNK || UNK method\tREL\tmethod used for task\n", "0.853682\tUNK method || UNK\tREL\tmethod used for task\n", "0.853607\thybrid method || complex\tREL\tmethod used for task\n", "0.853444\tnonlinear UNK || UNK\tREL\tmethod used for task\n", "0.853444\tnonlinear UNK || UNK\tREL\tmethod used for task\n", "0.853444\tnonlinear UNK || UNK\tREL\tmethod used for task\n", "0.853444\tnonlinear UNK || UNK\tREL\tmethod used for task\n", "0.853444\tnonlinear UNK || UNK\tREL\tmethod used for task\n", "0.853241\tnonlinear optimization || input\tREL\tmethod used for task\n", "0.853017\texperimental UNK || UNK detection\tREL\tmethod used for task\n", "0.852627\tdata UNK || data model\tREL\tmethod used for task\n", "0.851992\texperimental analysis || UNK models\tREL\tmethod used for task\n", "0.851870\toptimization technique || UNK\tREL\tmethod used for task\n", "0.851167\tUNK UNK || ontology construction\tREL\tmethod used for task\n", "0.851167\tUNK UNK || ontology construction\tREL\tmethod used for task\n", "0.851167\tUNK UNK || ontology construction\tREL\tmethod used for task\n", "0.851167\tUNK UNK || ontology construction\tREL\tmethod used for task\n", "0.851167\tUNK UNK || ontology construction\tREL\tmethod used for task\n", "0.851167\tUNK UNK || ontology construction\tREL\tmethod used for task\n", "0.849504\tmodeling UNK || UNK\tREL\tmethod used for task\n", "0.849504\tUNK modeling || UNK\tREL\tmethod used for task\n", "0.849504\tUNK UNK || UNK modeling\tREL\tmethod used for task\n", "0.849504\tUNK UNK || UNK modeling\tREL\tmethod used for task\n", "0.849504\tmodeling UNK || UNK\tREL\tmethod used for task\n", "0.849504\tUNK modeling || UNK\tREL\tmethod used for task\n", "0.849504\tUNK UNK || UNK modeling\tREL\tmethod used for task\n", "0.849504\tmodeling UNK || UNK\tREL\tmethod used for task\n", "0.849504\tUNK modeling || UNK\tREL\tmethod used for task\n", "0.849504\tUNK modeling || UNK\tREL\tmethod used for task\n", "0.849504\tUNK modeling || UNK\tREL\tmethod used for task\n", "0.849504\tUNK UNK || UNK modeling\tREL\tmethod used for task\n", "0.849504\tUNK UNK || UNK modeling\tREL\tmethod used for task\n", "0.849504\tUNK modeling || UNK\tREL\tmethod used for task\n", "0.849074\tUNK UNK method || UNK data\tREL\tmethod used for task\n", "0.848590\tUNK regression models || data mining\tREL\tmethod used for task\n", "0.848560\tUNK UNK || local\tREL\tmethod used for task\n", "0.848560\tlocal UNK || UNK\tREL\tmethod used for task\n", "0.848560\tlocal UNK || UNK\tREL\tmethod used for task\n", "0.848560\tlocal UNK || UNK\tREL\tmethod used for task\n", "0.848560\tlocal UNK UNK || UNK\tREL\tmethod used for task\n", "0.848560\tUNK UNK || local\tREL\tmethod used for task\n", "0.847563\trepresentation model || design\tREL\tmethod used for task\n", "0.846107\tUNK regression model || data mining\tREL\tmethod used for task\n", "0.842890\toptimal solution || UNK algorithm\tREL\tmethod used for task\n", "0.842695\tUNK UNK UNK || chain\tREL\tmethod used for task\n", "0.842469\toptimal shape || UNK\tREL\tmethod used for task\n", "0.841614\tUNK shape model || UNK\tREL\tmethod used for task\n", "0.841614\tshape UNK || UNK UNK model\tREL\tmethod used for task\n", "0.841614\tshape model || UNK\tREL\tmethod used for task\n", "0.839305\tshape models || UNK data\tREL\tmethod used for task\n", "0.838494\tnetwork design problem || UNK\tREL\tmethod used for task\n", "0.837771\tproduct model || design\tREL\tmethod used for task\n", "0.836611\tUNK UNK || UNK UNK theory\tREL\tmethod used for task\n", "0.836611\tUNK UNK theory || UNK\tREL\tmethod used for task\n", "0.836611\tUNK UNK theory || UNK\tREL\tmethod used for task\n", "0.836611\tUNK theory || UNK\tREL\tmethod used for task\n", "0.836611\tUNK UNK || UNK theory\tREL\tmethod used for task\n", "0.836611\tUNK UNK || UNK theory\tREL\tmethod used for task\n", "0.836611\tUNK UNK || UNK theory\tREL\tmethod used for task\n", "0.836344\tUNK systems || UNK problem\tREL\tmethod used for task\n", "0.834523\tUNK problem || computational\tREL\tmethod used for task\n", "0.834523\tcomputational UNK || problem\tREL\tmethod used for task\n", "0.833354\tmodel construction || UNK\tREL\tmethod used for task\n", "0.832792\tdata mining || data analysis\tREL\tmethod used for task\n", "0.830104\tUNK UNK || UNK chaos\tREL\tmethod used for task\n", "0.829606\tUNK clustering || UNK data analysis\tREL\tmethod used for task\n", "0.827977\tlinear UNK || UNK problem\tREL\tmethod used for task\n", "0.827033\tUNK UNK || UNK UNK design\tREL\tmethod used for task\n", "0.827033\tUNK UNK || UNK design\tREL\tmethod used for task\n", "0.827033\tUNK UNK || design\tREL\tmethod used for task\n", "0.827033\tUNK UNK || design\tREL\tmethod used for task\n", "0.827033\tUNK UNK || UNK design\tREL\tmethod used for task\n", "0.827033\tUNK UNK || UNK design\tREL\tmethod used for task\n", "0.827033\tUNK UNK || design\tREL\tmethod used for task\n", "0.827033\tUNK UNK || design\tREL\tmethod used for task\n", "0.827033\tUNK UNK || UNK design\tREL\tmethod used for task\n", "0.827033\tUNK UNK || design\tREL\tmethod used for task\n", "0.827033\tUNK UNK || design\tREL\tmethod used for task\n", "0.827033\tUNK UNK || design\tREL\tmethod used for task\n", "0.827033\tUNK UNK || design\tREL\tmethod used for task\n", "0.827033\tUNK UNK || design\tREL\tmethod used for task\n", "0.827033\tUNK UNK || UNK design\tREL\tmethod used for task\n", "0.827032\tUNK design || UNK\tREL\tmethod used for task\n", "0.827032\tUNK design || UNK\tREL\tmethod used for task\n", "0.827032\tdesign UNK || UNK\tREL\tmethod used for task\n", "0.827032\tdesign UNK || UNK\tREL\tmethod used for task\n", "0.827032\tdesign UNK || UNK\tREL\tmethod used for task\n", "0.827032\tdesign UNK || UNK\tREL\tmethod used for task\n", "0.827032\tUNK design || UNK\tREL\tmethod used for task\n", "0.827032\tUNK design || UNK\tREL\tmethod used for task\n", "0.827032\tdesign UNK || UNK\tREL\tmethod used for task\n", "0.827032\tdesign UNK || UNK\tREL\tmethod used for task\n", "0.827032\tdesign UNK || UNK\tREL\tmethod used for task\n", "0.827032\tdesign UNK || UNK\tREL\tmethod used for task\n", "0.827032\tdesign UNK || UNK\tREL\tmethod used for task\n", "0.827032\tUNK design || UNK\tREL\tmethod used for task\n", "0.827032\tUNK design || UNK\tREL\tmethod used for task\n", "0.827032\tUNK design || UNK\tREL\tmethod used for task\n", "0.827032\tdesign UNK || UNK\tREL\tmethod used for task\n", "0.827032\tUNK design || UNK\tREL\tmethod used for task\n", "0.827032\tUNK design || UNK\tREL\tmethod used for task\n", "0.827032\tdesign UNK || UNK\tREL\tmethod used for task\n", "0.827032\tdesign UNK || UNK\tREL\tmethod used for task\n", "0.827032\tdesign UNK || UNK\tREL\tmethod used for task\n", "0.825233\tUNK UNK networks || UNK analysis\tREL\tmethod used for task\n", "0.824216\tUNK method || UNK construction\tREL\tmethod used for task\n", "0.822423\tlocal search || solution\tREL\tmethod used for task\n", "0.822239\tUNK system || network\tREL\tmethod used for task\n", "0.822239\tsystem UNK || network\tREL\tmethod used for task\n", "0.822237\tshape models || shape\tREL\tmethod used for task\n", "0.821760\tUNK UNK || design data\tREL\tmethod used for task\n", "0.820596\tsearch UNK || search quality\tREL\tmethod used for task\n", "0.820596\tsearch UNK || search quality\tREL\tmethod used for task\n", "0.820596\tsearch UNK || search quality\tREL\tmethod used for task\n", "0.820596\tsearch quality || search\tREL\tmethod used for task\n", "0.820596\tsearch quality || search\tREL\tmethod used for task\n", "0.818770\texperimental UNK || UNK\tREL\tmethod used for task\n", "0.818770\texperimental UNK || UNK\tREL\tmethod used for task\n", "0.818770\texperimental UNK || UNK\tREL\tmethod used for task\n", "0.818770\texperimental UNK || UNK\tREL\tmethod used for task\n", "0.818770\texperimental UNK || UNK\tREL\tmethod used for task\n", "0.817900\tUNK systems || inverse\tREL\tmethod used for task\n", "0.817661\tUNK UNK || dynamic systems\tREL\tmethod used for task\n", "0.817364\tUNK parameters || UNK problem\tREL\tmethod used for task\n", "0.813302\texperimental data || UNK\tREL\tmethod used for task\n", "0.813302\tUNK UNK || experimental data\tREL\tmethod used for task\n", "0.813302\texperimental data || UNK\tREL\tmethod used for task\n", "0.813302\texperimental UNK || data\tREL\tmethod used for task\n", "0.811123\tUNK UNK model || customer\tREL\tmethod used for task\n", "0.810068\tapproximate solution || UNK networks\tREL\tmethod used for task\n", "0.809742\tpredictive models || UNK techniques\tREL\tmethod used for task\n", "0.809302\tcomputational cost || surrogate models\tREL\tmethod used for task\n", "0.807623\tproblem UNK || UNK analysis\tREL\tmethod used for task\n", "0.807413\tUNK UNK || detection performance\tREL\tmethod used for task\n", "0.807189\tUNK UNK || UNK materials\tREL\tmethod used for task\n", "0.807097\tlocal search || UNK size\tREL\tmethod used for task\n", "0.805169\tUNK systems || UNK algorithm\tREL\tmethod used for task\n", "0.803083\tcomputational UNK || UNK algorithm\tREL\tmethod used for task\n", "0.803083\tUNK algorithm || computational\tREL\tmethod used for task\n", "0.803083\tUNK algorithm || computational\tREL\tmethod used for task\n", "0.797932\tmodel UNK || solution method\tREL\tmethod used for task\n", "0.797932\tmodel UNK || solution method\tREL\tmethod used for task\n", "0.797365\tcomputational UNK || UNK search\tREL\tmethod used for task\n", "0.797365\tUNK search || computational\tREL\tmethod used for task\n", "0.796960\tUNK detection || nonlinear systems\tREL\tmethod used for task\n", "0.795598\tUNK algorithm || linear\tREL\tmethod used for task\n", "0.795590\tUNK analysis || topology\tREL\tmethod used for task\n", "0.794684\tcomputational UNK || domain\tREL\tmethod used for task\n", "0.792941\tUNK analysis || system parameters\tREL\tmethod used for task\n", "0.792394\toptimal solution || local\tREL\tmethod used for task\n", "0.791229\tUNK model || design space\tREL\tmethod used for task\n", "0.788207\tgenetic algorithm || solution technique\tREL\tmethod used for task\n", "0.787313\tUNK reliability || UNK reliability\tREL\tmethod used for task\n", "0.787313\tUNK reliability || UNK reliability\tREL\tmethod used for task\n", "0.786930\tspatial UNK || inverse relationship\tREL\tmethod used for task\n", "0.784502\tUNK UNK optimization || UNK planning\tREL\tmethod used for task\n", "0.784502\tUNK planning || optimization\tREL\tmethod used for task\n", "0.784130\tcase UNK || UNK model\tREL\tmethod used for task\n", "0.781642\tapproximate UNK || UNK\tREL\tmethod used for task\n", "0.781642\tUNK UNK || approximate\tREL\tmethod used for task\n", "0.781326\tdynamic simulation || UNK\tREL\tmethod used for task\n", "0.780309\tswarm intelligence || UNK neural network\tREL\tmethod used for task\n", "0.779790\tdetection UNK || detection\tREL\tmethod used for task\n", "0.779790\tUNK detection || detection\tREL\tmethod used for task\n", "0.779437\tcomputational method || UNK nonlinear analysis\tREL\tmethod used for task\n", "0.778773\tUNK cost || UNK UNK networks\tREL\tmethod used for task\n", "0.774374\tUNK space || ensemble\tREL\tmethod used for task\n", "0.771304\tlocal search || quality\tREL\tmethod used for task\n", "0.771136\tcase study || optimization model\tREL\tmethod used for task\n", "0.769587\tnumerical UNK || UNK algorithm\tREL\tmethod used for task\n", "0.769587\tnumerical UNK || UNK algorithm\tREL\tmethod used for task\n", "0.767054\tUNK UNK algorithm || simulation\tREL\tmethod used for task\n", "0.767054\tUNK UNK algorithm || simulation\tREL\tmethod used for task\n", "0.767054\tsimulation UNK || UNK algorithm\tREL\tmethod used for task\n", "0.767054\tsimulation UNK || UNK algorithm\tREL\tmethod used for task\n", "0.766693\tUNK system || UNK quality\tREL\tmethod used for task\n", "0.766693\tUNK system || UNK quality\tREL\tmethod used for task\n", "0.766691\toptimal UNK UNK || UNK systems\tREL\tmethod used for task\n", "0.765539\tUNK model || UNK systems\tREL\tmethod used for task\n", "0.765539\tUNK model || UNK systems\tREL\tmethod used for task\n", "0.765461\tUNK UNK || UNK performance\tREL\tmethod used for task\n", "0.765461\tUNK UNK || UNK performance\tREL\tmethod used for task\n", "0.765461\tUNK UNK || UNK performance\tREL\tmethod used for task\n", "0.765461\tUNK performance || UNK\tREL\tmethod used for task\n", "0.765461\tUNK UNK || UNK performance\tREL\tmethod used for task\n", "0.765461\tUNK performance || UNK\tREL\tmethod used for task\n", "0.765461\tperformance UNK || UNK\tREL\tmethod used for task\n", "0.765461\tperformance UNK || UNK\tREL\tmethod used for task\n", "0.765461\tUNK performance || UNK\tREL\tmethod used for task\n", "0.765461\tUNK UNK || performance\tREL\tmethod used for task\n", "0.765461\tperformance UNK || UNK\tREL\tmethod used for task\n", "0.765461\tUNK UNK || UNK performance\tREL\tmethod used for task\n", "0.765461\tUNK performance || UNK\tREL\tmethod used for task\n", "0.765461\tUNK performance || UNK\tREL\tmethod used for task\n", "0.765461\tUNK UNK || UNK performance\tREL\tmethod used for task\n", "0.765461\tUNK UNK || performance\tREL\tmethod used for task\n", "0.765461\tperformance UNK || UNK\tREL\tmethod used for task\n", "0.765461\tUNK UNK || performance\tREL\tmethod used for task\n", "0.765461\tUNK UNK UNK || UNK performance\tREL\tmethod used for task\n", "0.765461\tUNK performance || UNK\tREL\tmethod used for task\n", "0.765461\tUNK UNK || performance\tREL\tmethod used for task\n", "0.765461\tUNK UNK || performance\tREL\tmethod used for task\n", "0.765461\tUNK UNK || UNK performance\tREL\tmethod used for task\n", "0.765461\tUNK performance || UNK\tREL\tmethod used for task\n", "0.765461\tUNK performance || UNK\tREL\tmethod used for task\n", "0.765461\tUNK performance || UNK\tREL\tmethod used for task\n", "0.765461\tperformance UNK || UNK\tREL\tmethod used for task\n", "0.765461\tperformance UNK || UNK\tREL\tmethod used for task\n", "0.765461\tUNK performance || UNK\tREL\tmethod used for task\n", "0.765461\tperformance UNK || UNK\tREL\tmethod used for task\n", "0.765461\tperformance UNK || UNK\tREL\tmethod used for task\n", "0.765461\tperformance UNK || UNK\tREL\tmethod used for task\n", "0.765461\tUNK UNK || UNK performance\tREL\tmethod used for task\n", "0.765461\tUNK performance || UNK\tREL\tmethod used for task\n", "0.765461\tperformance UNK || UNK\tREL\tmethod used for task\n", "0.765461\tUNK performance || UNK\tREL\tmethod used for task\n", "0.765461\tUNK performance || UNK\tREL\tmethod used for task\n", "0.765461\tUNK UNK || UNK performance\tREL\tmethod used for task\n", "0.765461\tUNK UNK || UNK performance\tREL\tmethod used for task\n", "0.765461\tUNK UNK || UNK performance\tREL\tmethod used for task\n", "0.765461\tperformance UNK || UNK\tREL\tmethod used for task\n", "0.765461\tUNK UNK || UNK performance\tREL\tmethod used for task\n", "0.765461\tUNK UNK || UNK performance\tREL\tmethod used for task\n", "0.765461\tperformance UNK || UNK\tREL\tmethod used for task\n", "0.765461\tUNK UNK || UNK performance\tREL\tmethod used for task\n", "0.765461\tUNK UNK || UNK performance\tREL\tmethod used for task\n", "0.765461\tperformance UNK || UNK\tREL\tmethod used for task\n", "0.765461\tUNK UNK || UNK performance\tREL\tmethod used for task\n", "0.765461\tUNK UNK || performance\tREL\tmethod used for task\n", "0.765461\tUNK UNK || UNK performance\tREL\tmethod used for task\n", "0.765461\tUNK performance || UNK\tREL\tmethod used for task\n", "0.765461\tUNK UNK || UNK performance\tREL\tmethod used for task\n", "0.765461\tperformance UNK || UNK\tREL\tmethod used for task\n", "0.765461\tUNK UNK || UNK performance\tREL\tmethod used for task\n", "0.765461\tperformance UNK || UNK\tREL\tmethod used for task\n", "0.765461\tUNK UNK || UNK UNK performance\tREL\tmethod used for task\n", "0.765461\tUNK UNK || performance\tREL\tmethod used for task\n", "0.765461\tUNK UNK || performance\tREL\tmethod used for task\n", "0.765461\tUNK UNK || UNK performance\tREL\tmethod used for task\n", "0.765461\tperformance UNK || UNK\tREL\tmethod used for task\n", "0.765461\tUNK UNK || UNK performance\tREL\tmethod used for task\n", "0.765461\tUNK UNK || UNK performance\tREL\tmethod used for task\n", "0.765461\tperformance UNK || UNK\tREL\tmethod used for task\n", "0.765461\tUNK performance || UNK\tREL\tmethod used for task\n", "0.765461\tUNK performance || UNK\tREL\tmethod used for task\n", "0.765461\tUNK UNK || performance\tREL\tmethod used for task\n", "0.765461\tUNK performance || UNK\tREL\tmethod used for task\n", "0.765461\tUNK performance || UNK\tREL\tmethod used for task\n", "0.765461\tUNK UNK || performance\tREL\tmethod used for task\n", "0.765461\tperformance UNK || UNK\tREL\tmethod used for task\n", "0.765461\tperformance UNK || UNK\tREL\tmethod used for task\n", "0.765461\tUNK performance UNK || UNK\tREL\tmethod used for task\n", "0.765075\tformal UNK || UNK\tREL\tmethod used for task\n", "0.763154\tUNK UNK || computational model\tREL\tmethod used for task\n", "0.763154\tUNK UNK || computational model\tREL\tmethod used for task\n", "0.763154\tcomputational model || UNK\tREL\tmethod used for task\n", "0.763145\tUNK analysis || UNK domain\tREL\tmethod used for task\n", "0.760812\thybrid UNK || simulation\tREL\tmethod used for task\n", "0.760601\tUNK simulation || search\tREL\tmethod used for task\n", "0.759525\tUNK algorithm || UNK representation\tREL\tmethod used for task\n", "0.758859\tUNK data || UNK performance\tREL\tmethod used for task\n", "0.758859\tUNK data || UNK performance\tREL\tmethod used for task\n", "0.758466\tUNK data || formal\tREL\tmethod used for task\n", "0.758034\tUNK algorithm || UNK technique\tREL\tmethod used for task\n", "0.757856\tUNK problem || cost\tREL\tmethod used for task\n", "0.757856\tUNK problem || UNK cost\tREL\tmethod used for task\n", "0.756409\tmodel development || UNK\tREL\tmethod used for task\n", "0.755522\tbuilding UNK || UNK representation\tREL\tmethod used for task\n", "0.755379\tcomplex networks || UNK\tREL\tmethod used for task\n", "0.755379\tcomplex networks || UNK\tREL\tmethod used for task\n", "0.754615\tUNK UNK model || linear\tREL\tmethod used for task\n", "0.754411\tUNK meshes || UNK problem\tREL\tmethod used for task\n", "0.753644\tUNK UNK system || regression\tREL\tmethod used for task\n", "0.751989\tcomputational UNK || UNK UNK techniques\tREL\tmethod used for task\n", "0.751314\tcomputational UNK || UNK method\tREL\tmethod used for task\n", "0.751314\tcomputational method || UNK\tREL\tmethod used for task\n", "0.749635\tUNK language UNK || UNK\tREL\tmethod used for task\n", "0.749635\tUNK language UNK || UNK\tREL\tmethod used for task\n", "0.749635\tlanguage UNK || UNK\tREL\tmethod used for task\n", "0.749635\tUNK language UNK || UNK\tREL\tmethod used for task\n", "0.749635\tUNK language UNK || UNK\tREL\tmethod used for task\n", "0.749635\tUNK UNK || UNK language\tREL\tmethod used for task\n", "0.749635\tUNK language UNK || UNK\tREL\tmethod used for task\n", "0.749635\tUNK UNK || language\tREL\tmethod used for task\n", "0.747908\tUNK models || UNK regression models\tREL\tmethod used for task\n", "0.745085\tcomputational modeling || UNK\tREL\tmethod used for task\n", "0.744387\tdesign space || design\tREL\tmethod used for task\n", "0.740892\tUNK UNK model || UNK parameters\tREL\tmethod used for task\n", "0.740892\tUNK UNK model || UNK parameters\tREL\tmethod used for task\n", "0.740892\tUNK UNK || model parameters\tREL\tmethod used for task\n", "0.740892\tUNK UNK || model parameters\tREL\tmethod used for task\n", "0.740892\tUNK model || UNK parameters\tREL\tmethod used for task\n", "0.740892\tmodel parameters || UNK\tREL\tmethod used for task\n", "0.740600\tregression model || UNK model\tREL\tmethod used for task\n", "0.740514\tUNK ontology || UNK representation\tREL\tmethod used for task\n", "0.735935\tshape UNK || UNK performance\tREL\tmethod used for task\n", "0.735916\tUNK UNK model || fire\tREL\tmethod used for task\n", "0.733804\tUNK detection || UNK\tREL\tmethod used for task\n", "0.733804\tUNK detection || UNK\tREL\tmethod used for task\n", "0.733804\tUNK detection || UNK\tREL\tmethod used for task\n", "0.733804\tdetection UNK || UNK\tREL\tmethod used for task\n", "0.733804\tdetection UNK || UNK\tREL\tmethod used for task\n", "0.733804\tUNK detection || UNK\tREL\tmethod used for task\n", "0.733804\tdetection UNK || UNK\tREL\tmethod used for task\n", "0.733804\tUNK detection || UNK\tREL\tmethod used for task\n", "0.733804\tUNK detection || UNK\tREL\tmethod used for task\n", "0.733804\tUNK detection || UNK\tREL\tmethod used for task\n", "0.733804\tdetection UNK || UNK\tREL\tmethod used for task\n", "0.733804\tdetection UNK || UNK\tREL\tmethod used for task\n", "0.733804\tUNK detection || UNK\tREL\tmethod used for task\n", "0.733804\tUNK detection || UNK\tREL\tmethod used for task\n", "0.733804\tUNK detection || UNK\tREL\tmethod used for task\n", "0.733804\tUNK detection || UNK\tREL\tmethod used for task\n", "0.733804\tUNK UNK UNK UNK || UNK detection\tREL\tmethod used for task\n", "0.733804\tUNK UNK UNK || UNK detection\tREL\tmethod used for task\n", "0.733804\tUNK UNK || UNK detection\tREL\tmethod used for task\n", "0.733804\tUNK UNK || UNK detection\tREL\tmethod used for task\n", "0.733804\tUNK UNK || detection\tREL\tmethod used for task\n", "0.733804\tUNK UNK || UNK detection\tREL\tmethod used for task\n", "0.733804\tUNK UNK || UNK detection\tREL\tmethod used for task\n", "0.733804\tUNK UNK || UNK detection\tREL\tmethod used for task\n", "0.733804\tUNK UNK || UNK detection\tREL\tmethod used for task\n", "0.733804\tUNK UNK || UNK detection\tREL\tmethod used for task\n", "0.733804\tUNK UNK || UNK detection\tREL\tmethod used for task\n", "0.733804\tUNK UNK || UNK detection\tREL\tmethod used for task\n", "0.733804\tUNK UNK UNK || UNK detection\tREL\tmethod used for task\n", "0.733804\tUNK UNK || UNK detection\tREL\tmethod used for task\n", "0.733804\tUNK UNK || detection\tREL\tmethod used for task\n", "0.733804\tUNK UNK || UNK detection\tREL\tmethod used for task\n", "0.733804\tUNK UNK UNK || UNK UNK detection\tREL\tmethod used for task\n", "0.733804\tUNK UNK || UNK UNK detection\tREL\tmethod used for task\n", "0.733804\tUNK UNK || UNK detection\tREL\tmethod used for task\n", "0.733530\tlanguage model || UNK neural network\tREL\tmethod used for task\n", "0.732209\tUNK analysis || UNK models\tREL\tmethod used for task\n", "0.731013\tUNK reliability || UNK data\tREL\tmethod used for task\n", "0.729697\tUNK analysis || UNK optimal\tREL\tmethod used for task\n", "0.728999\tnumerical models || UNK\tREL\tmethod used for task\n", "0.728428\tUNK analysis || model\tREL\tmethod used for task\n", "0.728428\tUNK analysis || UNK model\tREL\tmethod used for task\n", "0.728428\tUNK analysis || UNK model\tREL\tmethod used for task\n", "0.728428\tUNK model || UNK analysis\tREL\tmethod used for task\n", "0.726630\tdetection UNK || UNK UNK data\tREL\tmethod used for task\n", "0.726630\tUNK detection || UNK data\tREL\tmethod used for task\n", "0.726212\tcomputational modeling || experimental analysis\tREL\tmethod used for task\n", "0.726179\tsimulation models || UNK\tREL\tmethod used for task\n", "0.726179\tsimulation models || UNK\tREL\tmethod used for task\n", "0.726179\tUNK UNK || simulation models\tREL\tmethod used for task\n", "0.725190\tUNK model || numerical\tREL\tmethod used for task\n", "0.725190\tnumerical model || UNK\tREL\tmethod used for task\n", "0.725190\tnumerical model || UNK\tREL\tmethod used for task\n", "0.723816\tnetwork UNK || UNK networks\tREL\tmethod used for task\n", "0.723739\tUNK UNK || wireless network\tREL\tmethod used for task\n", "0.722353\tUNK population || optimization\tREL\tmethod used for task\n", "0.722345\tsimulation UNK || UNK model\tREL\tmethod used for task\n", "0.722345\tsimulation model || UNK\tREL\tmethod used for task\n", "0.722345\tsimulation model || UNK\tREL\tmethod used for task\n", "0.721995\tgenetic optimization || input variables\tREL\tmethod used for task\n", "0.721372\tcase study || design optimization\tREL\tmethod used for task\n", "0.719074\tclustering method || case study\tREL\tmethod used for task\n", "0.716793\tUNK UNK algorithm || UNK cost\tREL\tmethod used for task\n", "0.716793\tUNK algorithm || cost\tREL\tmethod used for task\n", "0.716793\tUNK algorithm || UNK cost\tREL\tmethod used for task\n", "0.716793\tcost UNK || UNK algorithm\tREL\tmethod used for task\n", "0.716234\tUNK UNK analysis || UNK UNK techniques\tREL\tmethod used for task\n", "0.715499\tUNK analysis || UNK method\tREL\tmethod used for task\n", "0.715499\tanalysis method || UNK\tREL\tmethod used for task\n", "0.715499\tUNK analysis || UNK method\tREL\tmethod used for task\n", "0.715110\tnonlinear UNK || UNK analysis\tREL\tmethod used for task\n", "0.715009\tUNK UNK UNK systems || design\tREL\tmethod used for task\n", "0.714052\tUNK algorithm || neural network\tREL\tmethod used for task\n", "0.713910\tUNK representation || UNK model\tREL\tmethod used for task\n", "0.712985\tUNK meshes || UNK algorithm\tREL\tmethod used for task\n", "0.712985\tUNK algorithm || UNK meshes\tREL\tmethod used for task\n", "0.712243\tUNK UNK technique || UNK model\tREL\tmethod used for task\n", "0.712167\tUNK UNK method || numerical\tREL\tmethod used for task\n", "0.712167\tUNK method || numerical\tREL\tmethod used for task\n", "0.712167\tnumerical UNK || UNK method\tREL\tmethod used for task\n", "0.712167\tnumerical method || UNK\tREL\tmethod used for task\n", "0.711048\tcomputational intelligence || UNK\tREL\tmethod used for task\n", "0.710339\tUNK UNK || UNK planning problem\tREL\tmethod used for task\n", "0.710339\tUNK UNK || UNK planning problem\tREL\tmethod used for task\n", "0.710339\tUNK planning || problem\tREL\tmethod used for task\n", "0.710339\tUNK UNK problem || UNK planning\tREL\tmethod used for task\n", "0.709713\tUNK cost || hybrid\tREL\tmethod used for task\n", "0.709242\tUNK UNK || simulation method\tREL\tmethod used for task\n", "0.708721\tmodeling UNK || UNK analysis\tREL\tmethod used for task\n", "0.708721\tUNK modeling || UNK analysis\tREL\tmethod used for task\n", "0.704670\tUNK design || UNK development\tREL\tmethod used for task\n", "0.702612\tUNK models || product\tREL\tmethod used for task\n", "0.702377\tUNK simulation || UNK modeling\tREL\tmethod used for task\n", "0.701845\tshape UNK || UNK detection\tREL\tmethod used for task\n", "0.700137\tUNK analysis || UNK shape models\tREL\tmethod used for task\n", "0.699706\tUNK UNK UNK || UNK network problem\tREL\tmethod used for task\n", "0.698759\tUNK problem || UNK variables\tREL\tmethod used for task\n", "0.697936\tthermal UNK || UNK design\tREL\tmethod used for task\n", "0.695393\tUNK algorithm || experimental study\tREL\tmethod used for task\n", "0.695393\texperimental study || UNK algorithm\tREL\tmethod used for task\n", "0.694489\tsolution UNK || problem\tREL\tmethod used for task\n", "0.694489\tsolution UNK || UNK problem\tREL\tmethod used for task\n", "0.694489\tUNK UNK problem || solution\tREL\tmethod used for task\n", "0.693980\tsupply UNK || product\tREL\tmethod used for task\n", "0.693980\tsupply UNK || product\tREL\tmethod used for task\n", "0.693593\tUNK modeling || UNK representation\tREL\tmethod used for task\n", "0.691859\tUNK UNK || modeling technique\tREL\tmethod used for task\n", "0.691859\tUNK modeling || UNK technique\tREL\tmethod used for task\n", "0.691859\tUNK modeling || UNK technique\tREL\tmethod used for task\n", "0.690136\tUNK networks || data size\tREL\tmethod used for task\n", "0.688194\tUNK analysis || UNK theory\tREL\tmethod used for task\n", "0.687219\tUNK UNK || design parameters\tREL\tmethod used for task\n", "0.686338\tsimulation test || system performance\tREL\tmethod used for task\n", "0.686338\tsystem performance UNK || simulation test\tREL\tmethod used for task\n", "0.685848\texperimental UNK || thermal\tREL\tmethod used for task\n", "0.683567\tUNK UNK || network topology\tREL\tmethod used for task\n", "0.682869\tperformance analysis || computational models\tREL\tmethod used for task\n", "0.682729\tUNK planning || dynamic\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682130\tUNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.682129\tUNK UNK UNK || UNK\tREL\tmethod used for task\n", "0.675094\tsatisfaction UNK || UNK\tREL\tmethod used for task\n", "0.675094\tUNK satisfaction || UNK\tREL\tmethod used for task\n", "0.675094\tUNK satisfaction || UNK\tREL\tmethod used for task\n", "0.675094\tUNK satisfaction || UNK\tREL\tmethod used for task\n", "0.675094\tUNK satisfaction || UNK\tREL\tmethod used for task\n", "0.675094\tUNK UNK || satisfaction\tREL\tmethod used for task\n", "0.675094\tUNK UNK || satisfaction\tREL\tmethod used for task\n", "0.675094\tUNK UNK || satisfaction\tREL\tmethod used for task\n", "0.675034\tsearch space || UNK\tREL\tmethod used for task\n", "0.675034\tsearch space || UNK\tREL\tmethod used for task\n", "0.675034\tsearch space || UNK\tREL\tmethod used for task\n", "0.675034\tsearch space || UNK\tREL\tmethod used for task\n", "0.675034\tsearch space || UNK\tREL\tmethod used for task\n", "0.675034\tsearch space || UNK\tREL\tmethod used for task\n", "0.675034\tUNK UNK || search space\tREL\tmethod used for task\n", "0.675034\tUNK UNK || search space\tREL\tmethod used for task\n", "0.675034\tUNK UNK || search space\tREL\tmethod used for task\n", "0.674181\tUNK UNK || UNK UNK data\tREL\tmethod used for task\n", "0.674181\tUNK UNK UNK || UNK UNK data\tREL\tmethod used for task\n", "0.674181\tUNK UNK UNK || UNK data\tREL\tmethod used for task\n", "0.674181\tUNK UNK || UNK UNK data\tREL\tmethod used for task\n", "0.674181\tUNK UNK UNK || UNK data\tREL\tmethod used for task\n", "0.674181\tUNK UNK UNK || data\tREL\tmethod used for task\n", "0.674181\tUNK data UNK || UNK\tREL\tmethod used for task\n", "0.674181\tUNK data UNK || UNK\tREL\tmethod used for task\n", "0.674181\tUNK UNK data || UNK\tREL\tmethod used for task\n", "0.674181\tUNK UNK UNK || data\tREL\tmethod used for task\n", "0.674181\tUNK data || UNK\tREL\tmethod used for task\n", "0.674181\tUNK UNK || UNK data\tREL\tmethod used for task\n", "0.674181\tUNK data || UNK\tREL\tmethod used for task\n", "0.674181\tdata UNK || UNK\tREL\tmethod used for task\n", "0.674181\tdata UNK || UNK\tREL\tmethod used for task\n", "0.674181\tUNK data || UNK\tREL\tmethod used for task\n", "0.674181\tUNK UNK || UNK data\tREL\tmethod used for task\n", "0.674181\tUNK data || UNK\tREL\tmethod used for task\n", "0.674181\tUNK UNK || UNK data\tREL\tmethod used for task\n", "0.674181\tdata UNK || UNK\tREL\tmethod used for task\n", "0.674181\tUNK UNK || UNK data\tREL\tmethod used for task\n", "0.674181\tUNK data || UNK\tREL\tmethod used for task\n", "0.674181\tUNK data || UNK\tREL\tmethod used for task\n", "0.674181\tUNK data || UNK\tREL\tmethod used for task\n", "0.674181\tUNK data || UNK\tREL\tmethod used for task\n", "0.674181\tUNK data || UNK\tREL\tmethod used for task\n", "0.674181\tUNK UNK || UNK data\tREL\tmethod used for task\n", "0.674181\tUNK data || UNK\tREL\tmethod used for task\n", "0.674181\tUNK data || UNK\tREL\tmethod used for task\n", "0.674181\tUNK data || UNK\tREL\tmethod used for task\n", "0.674181\tUNK data || UNK\tREL\tmethod used for task\n", "0.674181\tUNK UNK || UNK data\tREL\tmethod used for task\n", "0.674181\tdata UNK || UNK\tREL\tmethod used for task\n", "0.674181\tdata UNK || UNK\tREL\tmethod used for task\n", "0.674181\tUNK UNK || UNK data\tREL\tmethod used for task\n", "0.674181\tUNK data || UNK\tREL\tmethod used for task\n", "0.674181\tUNK UNK || UNK data\tREL\tmethod used for task\n", "0.674181\tUNK UNK || data\tREL\tmethod used for task\n", "0.674181\tUNK UNK || data\tREL\tmethod used for task\n", "0.674181\tdata UNK || UNK\tREL\tmethod used for task\n", "0.674181\tdata UNK || UNK\tREL\tmethod used for task\n", "0.674181\tUNK UNK || data\tREL\tmethod used for task\n", "0.674181\tUNK UNK || data\tREL\tmethod used for task\n", "0.674181\tUNK UNK || data\tREL\tmethod used for task\n", "0.674181\tUNK UNK || data\tREL\tmethod used for task\n", "0.674181\tdata UNK || UNK\tREL\tmethod used for task\n", "0.674181\tdata UNK || UNK\tREL\tmethod used for task\n", "0.674181\tdata UNK || UNK\tREL\tmethod used for task\n", "0.674181\tUNK UNK || UNK data\tREL\tmethod used for task\n", "0.674181\tUNK UNK || data\tREL\tmethod used for task\n", "0.674181\tUNK UNK || UNK data\tREL\tmethod used for task\n", "0.674181\tdata UNK || UNK\tREL\tmethod used for task\n", "0.674181\tUNK data || UNK\tREL\tmethod used for task\n", "0.674181\tUNK data || UNK\tREL\tmethod used for task\n", "0.674181\tUNK data || UNK\tREL\tmethod used for task\n", "0.674181\tUNK data || UNK\tREL\tmethod used for task\n", "0.674181\tUNK UNK || UNK data\tREL\tmethod used for task\n", "0.674181\tdata UNK || UNK\tREL\tmethod used for task\n", "0.674181\tUNK data || UNK\tREL\tmethod used for task\n", "0.674181\tUNK data || UNK\tREL\tmethod used for task\n", "0.674181\tdata UNK || UNK\tREL\tmethod used for task\n", "0.674181\tUNK UNK || UNK data\tREL\tmethod used for task\n", "0.674181\tdata UNK || UNK\tREL\tmethod used for task\n", "0.674181\tUNK UNK || data\tREL\tmethod used for task\n", "0.674181\tdata UNK || UNK\tREL\tmethod used for task\n", "0.674181\tdata UNK || UNK\tREL\tmethod used for task\n", "0.674181\tUNK UNK || UNK data\tREL\tmethod used for task\n", "0.674181\tdata UNK || UNK\tREL\tmethod used for task\n", "0.674181\tdata UNK || UNK\tREL\tmethod used for task\n", "0.674181\tUNK data || UNK\tREL\tmethod used for task\n", "0.674181\tUNK UNK || UNK data\tREL\tmethod used for task\n", "0.674181\tUNK UNK || data\tREL\tmethod used for task\n", "0.674181\tUNK UNK || UNK data\tREL\tmethod used for task\n", "0.674181\tdata UNK || UNK\tREL\tmethod used for task\n", "0.674181\tUNK UNK || UNK data\tREL\tmethod used for task\n", "0.674181\tdata UNK || UNK\tREL\tmethod used for task\n", "0.674181\tUNK data || UNK\tREL\tmethod used for task\n", "0.674181\tUNK data || UNK\tREL\tmethod used for task\n", "0.674181\tUNK data || UNK\tREL\tmethod used for task\n", "0.674181\tUNK data || UNK\tREL\tmethod used for task\n", "0.674181\tUNK data || UNK\tREL\tmethod used for task\n", "0.674181\tUNK UNK || UNK data\tREL\tmethod used for task\n", "0.674181\tUNK UNK || data\tREL\tmethod used for task\n", "0.674181\tUNK UNK || UNK data\tREL\tmethod used for task\n", "0.674181\tdata UNK || UNK\tREL\tmethod used for task\n", "0.674181\tdata UNK || UNK\tREL\tmethod used for task\n", "0.674181\tUNK UNK || data\tREL\tmethod used for task\n", "0.674181\tUNK data || UNK\tREL\tmethod used for task\n", "0.674181\tUNK UNK || data\tREL\tmethod used for task\n", "0.674181\tUNK UNK || UNK data\tREL\tmethod used for task\n", "0.674181\tUNK UNK || UNK data\tREL\tmethod used for task\n", "0.674181\tUNK UNK || UNK data\tREL\tmethod used for task\n", "0.674181\tUNK UNK || UNK data\tREL\tmethod used for task\n", "0.674181\tUNK UNK || UNK data\tREL\tmethod used for task\n", "0.674181\tUNK UNK || UNK data\tREL\tmethod used for task\n", "0.674181\tUNK UNK || UNK data\tREL\tmethod used for task\n", "0.674181\tdata UNK || UNK\tREL\tmethod used for task\n", "0.674181\tUNK UNK || UNK data\tREL\tmethod used for task\n", "0.674181\tUNK UNK || UNK data\tREL\tmethod used for task\n", "0.674181\tUNK UNK || UNK data\tREL\tmethod used for task\n", "0.674181\tUNK UNK || UNK data\tREL\tmethod used for task\n", "0.674181\tUNK UNK || UNK data\tREL\tmethod used for task\n", "0.674181\tUNK UNK || UNK data\tREL\tmethod used for task\n", "0.674181\tUNK UNK || UNK data\tREL\tmethod used for task\n", "0.674181\tUNK UNK || UNK data\tREL\tmethod used for task\n", "0.674181\tUNK UNK || data\tREL\tmethod used for task\n", "0.674181\tUNK UNK || UNK data\tREL\tmethod used for task\n", "0.674181\tdata UNK || UNK\tREL\tmethod used for task\n", "0.674181\tUNK UNK || data\tREL\tmethod used for task\n", "0.673312\tUNK analysis || design\tREL\tmethod used for task\n", "0.673178\tshell UNK || UNK surface\tREL\tmethod used for task\n", "0.673178\tUNK surface UNK || shell\tREL\tmethod used for task\n", "0.673178\tUNK surface UNK || shell\tREL\tmethod used for task\n", "0.672519\tproblem size || UNK\tREL\tmethod used for task\n", "0.672160\tdynamic UNK || UNK surface\tREL\tmethod used for task\n", "0.669714\tnumerical UNK || design\tREL\tmethod used for task\n", "0.668056\toptimal UNK || cost\tREL\tmethod used for task\n", "0.668056\toptimal UNK || UNK cost\tREL\tmethod used for task\n", "0.666862\texperimental data || UNK parameters\tREL\tmethod used for task\n", "0.666629\tcost UNK || UNK model\tREL\tmethod used for task\n", "0.666131\tdata UNK || UNK data\tREL\tmethod used for task\n", "0.666131\tdata UNK || UNK data\tREL\tmethod used for task\n", "0.666131\tUNK data || UNK data\tREL\tmethod used for task\n", "0.666131\tdata UNK || data\tREL\tmethod used for task\n", "0.666131\tdata UNK || data\tREL\tmethod used for task\n", "0.666131\tUNK data || UNK data\tREL\tmethod used for task\n", "0.666131\tUNK data || UNK data\tREL\tmethod used for task\n", "0.666131\tUNK data || UNK UNK data\tREL\tmethod used for task\n", "0.666131\tUNK data || UNK data\tREL\tmethod used for task\n", "0.666131\tUNK data || UNK data\tREL\tmethod used for task\n", "0.666131\tUNK data UNK || UNK data\tREL\tmethod used for task\n", "0.666131\tUNK data || data\tREL\tmethod used for task\n", "0.666078\tdynamic UNK || solution\tREL\tmethod used for task\n", "0.664783\tUNK UNK algorithm || UNK planning\tREL\tmethod used for task\n", "0.663631\tneural network || UNK model\tREL\tmethod used for task\n", "0.663631\tUNK model || UNK UNK UNK neural network\tREL\tmethod used for task\n", "0.660718\texperimental UNK || UNK analysis\tREL\tmethod used for task\n", "0.660718\texperimental analysis || UNK\tREL\tmethod used for task\n", "0.652507\texperimental data || UNK analysis\tREL\tmethod used for task\n", "0.648800\texperimental data || numerical\tREL\tmethod used for task\n", "0.647676\tUNK algorithm || solution\tREL\tmethod used for task\n", "0.647676\tUNK UNK || solution algorithm\tREL\tmethod used for task\n", "0.646952\tUNK UNK || shape\tREL\tmethod used for task\n", "0.646952\tUNK UNK || shape\tREL\tmethod used for task\n", "0.646952\tUNK UNK || UNK shape\tREL\tmethod used for task\n", "0.646952\tshape UNK || UNK\tREL\tmethod used for task\n", "0.646952\tUNK UNK || shape\tREL\tmethod used for task\n", "0.646952\tUNK UNK || UNK shape\tREL\tmethod used for task\n", "0.646952\tshape UNK || UNK\tREL\tmethod used for task\n", "0.646952\tUNK shape || UNK\tREL\tmethod used for task\n", "0.646952\tshape UNK || UNK\tREL\tmethod used for task\n", "0.646952\tshape UNK || UNK\tREL\tmethod used for task\n", "0.646952\tUNK UNK || shape\tREL\tmethod used for task\n", "0.646952\tUNK UNK || shape\tREL\tmethod used for task\n", "0.646952\tUNK UNK || shape\tREL\tmethod used for task\n", "0.646952\tshape UNK || UNK\tREL\tmethod used for task\n", "0.646952\tshape UNK || UNK\tREL\tmethod used for task\n", "0.646952\tUNK UNK || UNK shape\tREL\tmethod used for task\n", "0.645411\tUNK UNK || hybrid network\tREL\tmethod used for task\n", "0.643767\tperformance analysis || UNK detection\tREL\tmethod used for task\n", "0.643091\tsensitivity analysis || shape optimization\tREL\tmethod used for task\n", "0.640402\tproduct design || UNK\tREL\tmethod used for task\n", "0.640402\tproduct design || UNK\tREL\tmethod used for task\n", "0.638905\tartificial UNK || UNK\tREL\tmethod used for task\n", "0.633451\tgenetic UNK || UNK regression\tREL\tmethod used for task\n", "0.632968\tUNK construction || UNK\tREL\tmethod used for task\n", "0.631514\tfrequency domain || UNK\tREL\tmethod used for task\n", "0.629760\tUNK space || model\tREL\tmethod used for task\n", "0.629760\tUNK model || UNK space\tREL\tmethod used for task\n", "0.629760\tUNK space model || UNK\tREL\tmethod used for task\n", "0.629422\tUNK UNK UNK neural network || chain\tREL\tmethod used for task\n", "0.628248\tcomputational UNK || UNK performance\tREL\tmethod used for task\n", "0.624162\tsize UNK || UNK algorithm\tREL\tmethod used for task\n", "0.619295\tUNK model || artificial neural network\tREL\tmethod used for task\n", "0.619295\tUNK UNK || artificial neural network model\tREL\tmethod used for task\n", "0.615200\tsensitivity analysis || UNK networks\tREL\tmethod used for task\n", "0.614970\tUNK models || UNK planning\tREL\tmethod used for task\n", "0.614737\treliability analysis || UNK reliability\tREL\tmethod used for task\n", "0.614167\tnonlinear UNK || UNK UNK space\tREL\tmethod used for task\n", "0.613940\tgenetic algorithm || solution quality\tREL\tmethod used for task\n", "0.613940\tgenetic algorithm || solution quality\tREL\tmethod used for task\n", "0.610414\tUNK model || UNK planning\tREL\tmethod used for task\n", "0.609334\tpredictive models || UNK\tREL\tmethod used for task\n", "0.609334\tUNK UNK || predictive models\tREL\tmethod used for task\n", "0.608436\tcase study || UNK UNK system\tREL\tmethod used for task\n", "0.604755\tmodel UNK || predictive\tREL\tmethod used for task\n", "0.604755\tpredictive model || UNK\tREL\tmethod used for task\n", "0.603680\tUNK algorithm || wind\tREL\tmethod used for task\n", "0.602322\tUNK system || automatic analysis\tREL\tmethod used for task\n", "0.600690\tlanguage models || UNK quality\tREL\tmethod used for task\n", "0.598852\tsurface UNK || UNK model\tREL\tmethod used for task\n", "0.598188\tUNK UNK || UNK network model\tREL\tmethod used for task\n", "0.598188\tmodel UNK || network\tREL\tmethod used for task\n", "0.596933\tcustomer UNK || UNK\tREL\tmethod used for task\n", "0.596933\tUNK UNK UNK || customer\tREL\tmethod used for task\n", "0.595678\tUNK concrete UNK || fire\tREL\tmethod used for task\n", "0.595678\tUNK concrete UNK || fire\tREL\tmethod used for task\n", "0.594287\tlocal UNK UNK || artificial neural network\tREL\tmethod used for task\n", "0.593786\toptimal solution || UNK\tREL\tmethod used for task\n", "0.593786\toptimal solution || UNK\tREL\tmethod used for task\n", "0.593786\toptimal UNK || solution\tREL\tmethod used for task\n", "0.593786\tUNK UNK || optimal solution\tREL\tmethod used for task\n", "0.593786\tUNK UNK || optimal solution\tREL\tmethod used for task\n", "0.593786\tUNK optimal UNK || solution\tREL\tmethod used for task\n", "0.593786\toptimal UNK || UNK solution\tREL\tmethod used for task\n", "0.592942\tUNK UNK || supply network\tREL\tmethod used for task\n", "0.592235\tUNK solution || UNK model\tREL\tmethod used for task\n", "0.592235\tUNK model || UNK solution\tREL\tmethod used for task\n", "0.589404\tUNK data || UNK network model\tREL\tmethod used for task\n", "0.589148\tcustomer satisfaction || UNK\tREL\tmethod used for task\n", "0.586314\tUNK algorithm || UNK population\tREL\tmethod used for task\n", "0.585881\tgenetic algorithm || input variables\tREL\tmethod used for task\n", "0.584972\toptimal solution || UNK data\tREL\tmethod used for task\n", "0.584510\tperformance UNK || UNK analysis\tREL\tmethod used for task\n", "0.583469\tdata model || automatic detection\tREL\tmethod used for task\n", "0.583283\tUNK surface UNK || UNK method\tREL\tmethod used for task\n", "0.582710\tnumerical simulation || UNK algorithm\tREL\tmethod used for task\n", "0.576591\tUNK solution || UNK UNK method\tREL\tmethod used for task\n", "0.576591\tUNK UNK || solution method\tREL\tmethod used for task\n", "0.576591\tUNK UNK || solution method\tREL\tmethod used for task\n", "0.575639\tUNK performance || data analysis\tREL\tmethod used for task\n", "0.574842\tUNK swarm optimization || solution quality\tREL\tmethod used for task\n", "0.573162\tUNK text || quality data\tREL\tmethod used for task\n", "0.572816\tUNK study || UNK UNK problem\tREL\tmethod used for task\n", "0.572408\tUNK algorithm || quality\tREL\tmethod used for task\n", "0.568626\tarchitectural design || UNK\tREL\tmethod used for task\n", "0.567491\tUNK UNK || model size\tREL\tmethod used for task\n", "0.567491\tUNK size || UNK model\tREL\tmethod used for task\n", "0.567206\tUNK algorithm || computational cost\tREL\tmethod used for task\n", "0.566536\tdesign space || UNK\tREL\tmethod used for task\n", "0.562399\tsurface UNK || chain\tREL\tmethod used for task\n", "0.560610\tchain UNK || UNK variables\tREL\tmethod used for task\n", "0.560610\tchain UNK || UNK variables\tREL\tmethod used for task\n", "0.556082\tUNK case || UNK\tREL\tmethod used for task\n", "0.556082\tUNK case || UNK\tREL\tmethod used for task\n", "0.556082\tcase UNK || UNK\tREL\tmethod used for task\n", "0.556082\tUNK UNK || case\tREL\tmethod used for task\n", "0.556082\tUNK UNK || case\tREL\tmethod used for task\n", "0.556082\tUNK UNK UNK || case\tREL\tmethod used for task\n", "0.550564\tUNK UNK theory || UNK network\tREL\tmethod used for task\n", "0.549450\tUNK theory || UNK UNK variables\tREL\tmethod used for task\n", "0.547758\toptimal UNK || wind\tREL\tmethod used for task\n", "0.547206\tpractice UNK || UNK\tREL\tmethod used for task\n", "0.547074\tUNK case || UNK data\tREL\tmethod used for task\n", "0.543011\tUNK analysis || UNK detection\tREL\tmethod used for task\n", "0.543011\tUNK detection || UNK UNK analysis\tREL\tmethod used for task\n", "0.540575\tUNK study || dynamic\tREL\tmethod used for task\n", "0.536544\tUNK clustering || frequency analysis\tREL\tmethod used for task\n", "0.535425\tsimulation UNK || detection\tREL\tmethod used for task\n", "0.534253\tsurface UNK || UNK design\tREL\tmethod used for task\n", "0.533565\tnetwork design || UNK\tREL\tmethod used for task\n", "0.533396\tsimulation UNK || analysis models\tREL\tmethod used for task\n", "0.529633\tUNK UNK || UNK systems\tREL\tmethod used for task\n", "0.529633\tUNK UNK || systems\tREL\tmethod used for task\n", "0.529633\tUNK UNK || UNK systems\tREL\tmethod used for task\n", "0.529633\tUNK UNK || UNK systems\tREL\tmethod used for task\n", "0.529633\tUNK UNK || UNK systems\tREL\tmethod used for task\n", "0.529633\tUNK UNK || UNK systems\tREL\tmethod used for task\n", "0.529633\tUNK UNK || UNK systems\tREL\tmethod used for task\n", "0.529633\tUNK UNK || UNK systems\tREL\tmethod used for task\n", "0.529633\tUNK UNK || UNK systems\tREL\tmethod used for task\n", "0.529633\tUNK UNK || UNK systems\tREL\tmethod used for task\n", "0.529633\tUNK UNK || UNK systems\tREL\tmethod used for task\n", "0.529633\tUNK UNK || UNK systems\tREL\tmethod used for task\n", "0.529633\tUNK UNK || UNK systems\tREL\tmethod used for task\n", "0.529633\tUNK UNK || UNK systems\tREL\tmethod used for task\n", "0.529633\tUNK UNK || UNK systems\tREL\tmethod used for task\n", "0.529633\tUNK UNK || UNK systems\tREL\tmethod used for task\n", "0.529633\tUNK UNK || UNK UNK UNK systems\tREL\tmethod used for task\n", "0.529633\tUNK UNK || UNK UNK systems\tREL\tmethod used for task\n", "0.529633\tsystems UNK || UNK\tREL\tmethod used for task\n", "0.529633\tUNK systems || UNK\tREL\tmethod used for task\n", "0.529633\tUNK UNK UNK systems || UNK\tREL\tmethod used for task\n", "0.529633\tUNK systems || UNK\tREL\tmethod used for task\n", "0.529633\tUNK systems || UNK\tREL\tmethod used for task\n", "0.529633\tUNK UNK systems || UNK\tREL\tmethod used for task\n", "0.529633\tUNK systems || UNK\tREL\tmethod used for task\n", "0.529633\tUNK systems || UNK\tREL\tmethod used for task\n", "0.529633\tUNK systems || UNK\tREL\tmethod used for task\n", "0.529633\tUNK systems || UNK\tREL\tmethod used for task\n", "0.529633\tUNK systems || UNK\tREL\tmethod used for task\n", "0.526333\tUNK UNK || computational\tREL\tmethod used for task\n", "0.526333\tUNK UNK || computational\tREL\tmethod used for task\n", "0.526333\tUNK UNK || computational\tREL\tmethod used for task\n", "0.526333\tUNK UNK || computational\tREL\tmethod used for task\n", "0.526333\tcomputational UNK || UNK\tREL\tmethod used for task\n", "0.526333\tcomputational UNK || UNK\tREL\tmethod used for task\n", "0.526333\tcomputational UNK || UNK\tREL\tmethod used for task\n", "0.526333\tcomputational UNK || UNK\tREL\tmethod used for task\n", "0.526333\tcomputational UNK || UNK\tREL\tmethod used for task\n", "0.526333\tcomputational UNK || UNK\tREL\tmethod used for task\n", "0.526333\tcomputational UNK || UNK\tREL\tmethod used for task\n", "0.526333\tcomputational UNK || UNK\tREL\tmethod used for task\n", "0.526333\tUNK UNK UNK || computational\tREL\tmethod used for task\n", "0.526333\tcomputational UNK || UNK\tREL\tmethod used for task\n", "0.526333\tcomputational UNK || UNK\tREL\tmethod used for task\n", "0.526333\tcomputational UNK || UNK\tREL\tmethod used for task\n", "0.526333\tcomputational UNK || UNK\tREL\tmethod used for task\n", "0.526333\tcomputational UNK || UNK\tREL\tmethod used for task\n", "0.526333\tcomputational UNK || UNK\tREL\tmethod used for task\n", "0.526333\tcomputational UNK || UNK\tREL\tmethod used for task\n", "0.526333\tcomputational UNK || UNK\tREL\tmethod used for task\n", "0.526333\tcomputational UNK || UNK\tREL\tmethod used for task\n", "0.526333\tcomputational UNK || UNK\tREL\tmethod used for task\n", "0.522442\tUNK optimization || surface meshes\tREL\tmethod used for task\n", "0.520552\tdata UNK || UNK UNK systems\tREL\tmethod used for task\n", "0.520552\tUNK systems || data\tREL\tmethod used for task\n", "0.519689\tUNK analysis || UNK nonlinear analysis\tREL\tmethod used for task\n", "0.518804\tUNK quality || UNK models\tREL\tmethod used for task\n", "0.518804\tUNK models || UNK quality\tREL\tmethod used for task\n", "0.518216\tsearch space || computational\tREL\tmethod used for task\n", "0.517111\tUNK UNK || UNK development\tREL\tmethod used for task\n", "0.517111\tUNK UNK || UNK development\tREL\tmethod used for task\n", "0.517111\tUNK UNK || UNK development\tREL\tmethod used for task\n", "0.517111\tUNK UNK || UNK development\tREL\tmethod used for task\n", "0.517111\tUNK UNK || UNK development\tREL\tmethod used for task\n", "0.517111\tUNK development || UNK\tREL\tmethod used for task\n", "0.517111\tUNK UNK || development\tREL\tmethod used for task\n", "0.517111\tdevelopment UNK || UNK\tREL\tmethod used for task\n", "0.514732\tcomputational intelligence || UNK UNK analysis\tREL\tmethod used for task\n", "0.514686\tlinear UNK || UNK\tREL\tmethod used for task\n", "0.514686\tlinear UNK || UNK\tREL\tmethod used for task\n", "0.514686\tUNK UNK || linear\tREL\tmethod used for task\n", "0.514686\tUNK UNK || linear\tREL\tmethod used for task\n", "0.514686\tlinear UNK || UNK\tREL\tmethod used for task\n", "0.514686\tUNK UNK || linear\tREL\tmethod used for task\n", "0.514686\tlinear UNK || UNK\tREL\tmethod used for task\n", "0.514686\tlinear UNK || UNK\tREL\tmethod used for task\n", "0.514686\tUNK UNK || linear\tREL\tmethod used for task\n", "0.514686\tUNK UNK || linear\tREL\tmethod used for task\n", "0.514686\tlinear UNK || UNK\tREL\tmethod used for task\n", "0.514686\tUNK UNK || linear\tREL\tmethod used for task\n", "0.514686\tlinear UNK || UNK\tREL\tmethod used for task\n", "0.514686\tUNK UNK || linear\tREL\tmethod used for task\n", "0.514686\tUNK UNK || linear\tREL\tmethod used for task\n", "0.514686\tUNK UNK || linear\tREL\tmethod used for task\n", "0.514686\tUNK UNK || UNK UNK linear\tREL\tmethod used for task\n", "0.514686\tlinear UNK || UNK\tREL\tmethod used for task\n", "0.514645\tUNK UNK || automatic UNK method\tREL\tmethod used for task\n", "0.514010\tUNK model || UNK quality\tREL\tmethod used for task\n", "0.514010\tquality model || UNK\tREL\tmethod used for task\n", "0.514010\tUNK UNK || quality model\tREL\tmethod used for task\n", "0.513345\tUNK condition || UNK\tREL\tmethod used for task\n", "0.513345\tUNK condition || UNK\tREL\tmethod used for task\n", "0.510926\tspatial UNK || UNK model\tREL\tmethod used for task\n", "0.508013\tUNK development || UNK data\tREL\tmethod used for task\n", "0.504909\tdata quality || model\tREL\tmethod used for task\n", "0.502622\tsupply network || UNK network model\tREL\tmethod used for task\n", "0.502087\texperimental analysis || computational\tREL\tmethod used for task\n", "0.497909\tUNK quality || UNK UNK method\tREL\tmethod used for task\n", "0.497393\tcustomer satisfaction || customer\tREL\tmethod used for task\n", "0.496497\tUNK parameters || UNK\tREL\tmethod used for task\n", "0.496497\tUNK parameters || UNK\tREL\tmethod used for task\n", "0.496497\tUNK parameters || UNK\tREL\tmethod used for task\n", "0.496497\tUNK parameters || UNK\tREL\tmethod used for task\n", "0.496497\tUNK UNK || UNK UNK parameters\tREL\tmethod used for task\n", "0.496497\tUNK parameters || UNK\tREL\tmethod used for task\n", "0.496497\tUNK parameters || UNK\tREL\tmethod used for task\n", "0.496497\tUNK UNK || UNK parameters\tREL\tmethod used for task\n", "0.496497\tUNK UNK || UNK parameters\tREL\tmethod used for task\n", "0.496497\tUNK UNK || UNK parameters\tREL\tmethod used for task\n", "0.496117\tUNK UNK || UNK regression model\tREL\tmethod used for task\n", "0.496117\tUNK regression model || UNK\tREL\tmethod used for task\n", "0.494823\tUNK spatial UNK || UNK UNK method\tREL\tmethod used for task\n", "0.492073\tneural network || nonlinear systems\tREL\tmethod used for task\n", "0.487395\tUNK parameters || UNK data\tREL\tmethod used for task\n", "0.484343\tcomputational cost || modeling\tREL\tmethod used for task\n", "0.480519\tUNK UNK || UNK analysis\tREL\tmethod used for task\n", "0.480519\tUNK UNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK UNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK UNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK UNK || UNK UNK analysis\tREL\tmethod used for task\n", "0.480519\tUNK UNK UNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK UNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK UNK UNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK UNK || UNK analysis\tREL\tmethod used for task\n", "0.480519\tUNK UNK || UNK analysis\tREL\tmethod used for task\n", "0.480519\tUNK UNK || UNK analysis\tREL\tmethod used for task\n", "0.480519\tUNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK UNK || UNK analysis\tREL\tmethod used for task\n", "0.480519\tUNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK UNK || UNK analysis\tREL\tmethod used for task\n", "0.480519\tUNK UNK || UNK analysis\tREL\tmethod used for task\n", "0.480519\tUNK UNK || UNK analysis\tREL\tmethod used for task\n", "0.480519\tUNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK UNK || UNK analysis\tREL\tmethod used for task\n", "0.480519\tUNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK UNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK UNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK UNK || UNK analysis\tREL\tmethod used for task\n", "0.480519\tUNK UNK || UNK UNK UNK analysis\tREL\tmethod used for task\n", "0.480519\tUNK UNK || UNK analysis\tREL\tmethod used for task\n", "0.480519\tUNK UNK || UNK UNK analysis\tREL\tmethod used for task\n", "0.480519\tUNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK UNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK UNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK UNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK UNK || UNK analysis\tREL\tmethod used for task\n", "0.480519\tUNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK UNK || UNK analysis\tREL\tmethod used for task\n", "0.480519\tUNK UNK || UNK analysis\tREL\tmethod used for task\n", "0.480519\tUNK UNK || UNK analysis\tREL\tmethod used for task\n", "0.480519\tUNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK UNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK UNK || UNK analysis\tREL\tmethod used for task\n", "0.480519\tUNK UNK || UNK analysis\tREL\tmethod used for task\n", "0.480519\tUNK UNK || UNK analysis\tREL\tmethod used for task\n", "0.480519\tUNK UNK || analysis\tREL\tmethod used for task\n", "0.480519\tUNK UNK || UNK analysis\tREL\tmethod used for task\n", "0.480519\tUNK UNK || UNK analysis\tREL\tmethod used for task\n", "0.480519\tUNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK UNK || UNK analysis\tREL\tmethod used for task\n", "0.480519\tUNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK UNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK UNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK UNK || analysis\tREL\tmethod used for task\n", "0.480519\tUNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK UNK || UNK analysis\tREL\tmethod used for task\n", "0.480519\tUNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK UNK || UNK analysis\tREL\tmethod used for task\n", "0.480519\tUNK UNK || UNK analysis\tREL\tmethod used for task\n", "0.480519\tUNK UNK || UNK analysis\tREL\tmethod used for task\n", "0.480519\tUNK UNK || UNK analysis\tREL\tmethod used for task\n", "0.480519\tUNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK analysis || UNK\tREL\tmethod used for task\n", "0.480519\tUNK UNK UNK || UNK analysis\tREL\tmethod used for task\n", "0.480519\tUNK UNK UNK || UNK analysis\tREL\tmethod used for task\n", "0.480519\tUNK UNK UNK || UNK analysis\tREL\tmethod used for task\n", "0.480444\tUNK design || wind\tREL\tmethod used for task\n", "0.480023\tUNK UNK || regression method\tREL\tmethod used for task\n", "0.479759\tpso algorithm || solution quality\tREL\tmethod used for task\n", "0.476450\tnumerical UNK || UNK\tREL\tmethod used for task\n", "0.476450\tnumerical UNK || UNK\tREL\tmethod used for task\n", "0.476450\tnumerical UNK || UNK\tREL\tmethod used for task\n", "0.476450\tnumerical UNK || UNK\tREL\tmethod used for task\n", "0.476450\tnumerical UNK || UNK\tREL\tmethod used for task\n", "0.476450\tnumerical UNK || UNK\tREL\tmethod used for task\n", "0.476450\tnumerical UNK || UNK\tREL\tmethod used for task\n", "0.476450\tnumerical UNK || UNK\tREL\tmethod used for task\n", "0.476450\tnumerical UNK || UNK\tREL\tmethod used for task\n", "0.476450\tnumerical UNK || UNK\tREL\tmethod used for task\n", "0.476450\tnumerical UNK || UNK\tREL\tmethod used for task\n", "0.476024\tlinear UNK || solution algorithm\tREL\tmethod used for task\n", "0.474846\tUNK UNK problem || solution technique\tREL\tmethod used for task\n", "0.474559\tquality UNK || UNK shape model\tREL\tmethod used for task\n", "0.472902\tsimulation UNK || UNK\tREL\tmethod used for task\n", "0.472902\tUNK UNK || simulation\tREL\tmethod used for task\n", "0.472902\tsimulation UNK || UNK\tREL\tmethod used for task\n", "0.472902\tUNK UNK || simulation\tREL\tmethod used for task\n", "0.472902\tUNK UNK simulation || UNK\tREL\tmethod used for task\n", "0.472902\tUNK simulation || UNK\tREL\tmethod used for task\n", "0.472902\tUNK UNK || simulation\tREL\tmethod used for task\n", "0.471493\tUNK space || UNK performance\tREL\tmethod used for task\n", "0.471436\tUNK UNK || UNK data analysis\tREL\tmethod used for task\n", "0.471436\tUNK analysis || UNK data\tREL\tmethod used for task\n", "0.471436\tdata analysis || UNK\tREL\tmethod used for task\n", "0.471436\tdata analysis || UNK\tREL\tmethod used for task\n", "0.471436\tdata analysis || UNK\tREL\tmethod used for task\n", "0.471436\tUNK analysis || UNK UNK data\tREL\tmethod used for task\n", "0.471436\tdata analysis || UNK\tREL\tmethod used for task\n", "0.471436\tUNK analysis || data\tREL\tmethod used for task\n", "0.471436\tUNK UNK || data analysis\tREL\tmethod used for task\n", "0.471436\tUNK data || UNK analysis\tREL\tmethod used for task\n", "0.466383\tUNK neural network || detection\tREL\tmethod used for task\n", "0.466195\thistorical data || UNK models\tREL\tmethod used for task\n", "0.462528\tUNK UNK || UNK representation\tREL\tmethod used for task\n", "0.462528\tUNK UNK || UNK representation\tREL\tmethod used for task\n", "0.462528\tUNK representation || UNK\tREL\tmethod used for task\n", "0.462528\tUNK representation || UNK\tREL\tmethod used for task\n", "0.462528\tUNK UNK || UNK representation\tREL\tmethod used for task\n", "0.462528\tUNK UNK || UNK representation\tREL\tmethod used for task\n", "0.462528\tUNK representation || UNK\tREL\tmethod used for task\n", "0.462528\tUNK representation || UNK\tREL\tmethod used for task\n", "0.462528\tUNK representation || UNK\tREL\tmethod used for task\n", "0.462528\tUNK representation || UNK\tREL\tmethod used for task\n", "0.462528\tUNK UNK || UNK representation\tREL\tmethod used for task\n", "0.462528\tUNK UNK || UNK representation\tREL\tmethod used for task\n", "0.461421\tUNK model || historical data\tREL\tmethod used for task\n", "0.460640\tgenetic UNK || linear regression\tREL\tmethod used for task\n", "0.460503\tUNK UNK || UNK technique\tREL\tmethod used for task\n", "0.460503\tUNK UNK || UNK technique\tREL\tmethod used for task\n", "0.460503\tUNK UNK || UNK technique\tREL\tmethod used for task\n", "0.460503\tUNK technique || UNK\tREL\tmethod used for task\n", "0.455190\tUNK UNK || approximate solution\tREL\tmethod used for task\n", "0.455190\tapproximate solution || UNK\tREL\tmethod used for task\n", "0.454790\tUNK data || simulation data\tREL\tmethod used for task\n", "0.451469\tdata UNK || UNK technique\tREL\tmethod used for task\n", "0.450732\thybrid method || solution quality\tREL\tmethod used for task\n", "0.448826\tUNK engineering || UNK reliability\tREL\tmethod used for task\n", "0.448826\tUNK engineering || UNK reliability\tREL\tmethod used for task\n", "0.448826\tUNK engineering || UNK reliability\tREL\tmethod used for task\n", "0.445872\tboundary UNK || design\tREL\tmethod used for task\n", "0.445214\tUNK UNK || predictive performance\tREL\tmethod used for task\n", "0.444220\tUNK UNK || UNK product\tREL\tmethod used for task\n", "0.444220\tproduct UNK || UNK\tREL\tmethod used for task\n", "0.444220\tproduct UNK || UNK\tREL\tmethod used for task\n", "0.444220\tproduct UNK || UNK\tREL\tmethod used for task\n", "0.444220\tUNK product || UNK\tREL\tmethod used for task\n", "0.444220\tUNK product || UNK\tREL\tmethod used for task\n", "0.444220\tUNK UNK || product\tREL\tmethod used for task\n", "0.444220\tUNK UNK || product\tREL\tmethod used for task\n", "0.444220\tUNK UNK || product\tREL\tmethod used for task\n", "0.444220\tUNK UNK || product\tREL\tmethod used for task\n", "0.441304\tshape analysis || UNK\tREL\tmethod used for task\n", "0.439067\tcase study || UNK UNK problem\tREL\tmethod used for task\n", "0.439067\tcase study || UNK UNK problem\tREL\tmethod used for task\n", "0.438458\tnetwork performance || UNK\tREL\tmethod used for task\n", "0.431978\tUNK UNK || wireless sensor network\tREL\tmethod used for task\n", "0.431420\tUNK UNK theory || input\tREL\tmethod used for task\n", "0.431420\tUNK UNK theory || input\tREL\tmethod used for task\n", "0.423584\tUNK UNK || shape representation\tREL\tmethod used for task\n", "0.419870\tUNK UNK relations || UNK\tREL\tmethod used for task\n", "0.419870\tUNK UNK UNK || UNK UNK relations\tREL\tmethod used for task\n", "0.419870\tUNK UNK relations || UNK\tREL\tmethod used for task\n", "0.419870\tUNK UNK || UNK relations\tREL\tmethod used for task\n", "0.419870\tUNK UNK || UNK relations\tREL\tmethod used for task\n", "0.418109\tUNK condition || optimal solution\tREL\tmethod used for task\n", "0.410941\tcomputational cost || detection performance\tREL\tmethod used for task\n", "0.409799\tcomputational study || problem\tREL\tmethod used for task\n", "0.409572\tsize UNK || UNK concrete\tREL\tmethod used for task\n", "0.408144\tUNK UNK || UNK cost\tREL\tmethod used for task\n", "0.408144\tUNK UNK || UNK cost\tREL\tmethod used for task\n", "0.408144\tUNK UNK || UNK cost\tREL\tmethod used for task\n", "0.408144\tUNK UNK || UNK cost\tREL\tmethod used for task\n", "0.408144\tUNK UNK || UNK cost\tREL\tmethod used for task\n", "0.408144\tUNK UNK || UNK cost\tREL\tmethod used for task\n", "0.408144\tUNK UNK || UNK cost\tREL\tmethod used for task\n", "0.408144\tUNK UNK || UNK cost\tREL\tmethod used for task\n", "0.408144\tUNK UNK || UNK cost\tREL\tmethod used for task\n", "0.408144\tUNK UNK || UNK cost\tREL\tmethod used for task\n", "0.408144\tUNK UNK || UNK cost\tREL\tmethod used for task\n", "0.408144\tUNK UNK || UNK cost\tREL\tmethod used for task\n", "0.408144\tUNK UNK || UNK cost\tREL\tmethod used for task\n", "0.408144\tUNK UNK || UNK cost\tREL\tmethod used for task\n", "0.408144\tUNK UNK UNK || cost\tREL\tmethod used for task\n", "0.408144\tUNK UNK || cost\tREL\tmethod used for task\n", "0.408144\tUNK cost || UNK\tREL\tmethod used for task\n", "0.408144\tUNK UNK || cost\tREL\tmethod used for task\n", "0.408144\tUNK cost || UNK\tREL\tmethod used for task\n", "0.408144\tUNK cost || UNK\tREL\tmethod used for task\n", "0.408144\tUNK cost || UNK\tREL\tmethod used for task\n", "0.408144\tcost UNK || UNK\tREL\tmethod used for task\n", "0.408144\tUNK UNK || cost\tREL\tmethod used for task\n", "0.408144\tUNK UNK || cost\tREL\tmethod used for task\n", "0.408144\tcost UNK || UNK\tREL\tmethod used for task\n", "0.408144\tcost UNK || UNK\tREL\tmethod used for task\n", "0.408144\tcost UNK || UNK\tREL\tmethod used for task\n", "0.408144\tcost UNK || UNK\tREL\tmethod used for task\n", "0.408144\tcost UNK || UNK\tREL\tmethod used for task\n", "0.408144\tcost UNK || UNK\tREL\tmethod used for task\n", "0.408144\tcost UNK || UNK\tREL\tmethod used for task\n", "0.408144\tcost UNK || UNK\tREL\tmethod used for task\n", "0.408144\tUNK UNK || cost\tREL\tmethod used for task\n", "0.408144\tUNK UNK || cost\tREL\tmethod used for task\n", "0.408144\tUNK cost || UNK\tREL\tmethod used for task\n", "0.408144\tUNK UNK || cost\tREL\tmethod used for task\n", "0.408144\tUNK UNK || cost\tREL\tmethod used for task\n", "0.408144\tUNK cost || UNK\tREL\tmethod used for task\n", "0.408144\tUNK cost || UNK\tREL\tmethod used for task\n", "0.408144\tUNK UNK || cost\tREL\tmethod used for task\n", "0.408144\tUNK UNK || cost\tREL\tmethod used for task\n", "0.408144\tUNK UNK || cost\tREL\tmethod used for task\n", "0.408144\tUNK UNK || cost\tREL\tmethod used for task\n", "0.407390\tUNK space || analysis method\tREL\tmethod used for task\n", "0.406929\tnonlinear analysis || UNK space\tREL\tmethod used for task\n", "0.404897\tUNK UNK || UNK UNK neural network\tREL\tmethod used for task\n", "0.404897\tUNK UNK UNK || neural network\tREL\tmethod used for task\n", "0.404897\tUNK UNK UNK neural network || UNK\tREL\tmethod used for task\n", "0.404897\tUNK UNK UNK neural network || UNK\tREL\tmethod used for task\n", "0.404897\tUNK UNK UNK neural network || UNK\tREL\tmethod used for task\n", "0.404897\tUNK UNK UNK || neural network\tREL\tmethod used for task\n", "0.404897\tUNK UNK UNK neural network || UNK\tREL\tmethod used for task\n", "0.404897\tUNK UNK UNK neural network || UNK\tREL\tmethod used for task\n", "0.404897\tUNK UNK || UNK neural network\tREL\tmethod used for task\n", "0.404897\tneural network || UNK\tREL\tmethod used for task\n", "0.404897\tneural network || UNK\tREL\tmethod used for task\n", "0.404897\tneural network || UNK\tREL\tmethod used for task\n", "0.404897\tneural network || UNK\tREL\tmethod used for task\n", "0.404897\tneural network || UNK\tREL\tmethod used for task\n", "0.404897\tneural network || UNK\tREL\tmethod used for task\n", "0.404897\tneural network || UNK\tREL\tmethod used for task\n", "0.404897\tneural network || UNK\tREL\tmethod used for task\n", "0.404897\tneural network || UNK\tREL\tmethod used for task\n", "0.404897\tneural network || UNK\tREL\tmethod used for task\n", "0.404897\tneural network || UNK\tREL\tmethod used for task\n", "0.404897\tneural network || UNK\tREL\tmethod used for task\n", "0.404897\tneural network || UNK\tREL\tmethod used for task\n", "0.404897\tUNK neural network || UNK\tREL\tmethod used for task\n", "0.404897\tUNK UNK || neural network\tREL\tmethod used for task\n", "0.404897\tUNK neural network || UNK\tREL\tmethod used for task\n", "0.404897\tUNK UNK || neural network\tREL\tmethod used for task\n", "0.403640\tUNK UNK UNK || UNK meshes\tREL\tmethod used for task\n", "0.399377\tdata UNK || UNK cost\tREL\tmethod used for task\n", "0.399377\tcost UNK || UNK data\tREL\tmethod used for task\n", "0.397043\ttext UNK || case study\tREL\tmethod used for task\n", "0.391509\tUNK solution || UNK detection\tREL\tmethod used for task\n", "0.383836\tUNK UNK || sensor\tREL\tmethod used for task\n", "0.383836\tUNK UNK || sensor\tREL\tmethod used for task\n", "0.383836\tUNK UNK || sensor\tREL\tmethod used for task\n", "0.383836\tsensor UNK || UNK\tREL\tmethod used for task\n", "0.383836\tUNK UNK || sensor\tREL\tmethod used for task\n", "0.383836\tUNK UNK || sensor\tREL\tmethod used for task\n", "0.383836\tsensor UNK || UNK\tREL\tmethod used for task\n", "0.383836\tsensor UNK || UNK\tREL\tmethod used for task\n", "0.383836\tsensor UNK || UNK\tREL\tmethod used for task\n", "0.383836\tUNK UNK || sensor\tREL\tmethod used for task\n", "0.383836\tUNK UNK || sensor\tREL\tmethod used for task\n", "0.383482\texperimental study || UNK\tREL\tmethod used for task\n", "0.383482\tUNK UNK || experimental study\tREL\tmethod used for task\n", "0.383482\texperimental study || UNK\tREL\tmethod used for task\n", "0.383482\texperimental study || UNK\tREL\tmethod used for task\n", "0.383482\texperimental study || UNK\tREL\tmethod used for task\n", "0.383482\texperimental study || UNK\tREL\tmethod used for task\n", "0.382717\tengineering UNK || UNK\tREL\tmethod used for task\n", "0.382717\tUNK engineering || UNK\tREL\tmethod used for task\n", "0.381754\tUNK planning || UNK planning problem\tREL\tmethod used for task\n", "0.381138\tUNK algorithm || linear regression\tREL\tmethod used for task\n", "0.376915\tUNK UNK UNK || UNK complex\tREL\tmethod used for task\n", "0.376915\tcomplex UNK || UNK\tREL\tmethod used for task\n", "0.375260\tsensor data || UNK\tREL\tmethod used for task\n", "0.375260\tdata UNK || UNK sensor\tREL\tmethod used for task\n", "0.371396\tUNK systems || systems\tREL\tmethod used for task\n", "0.369717\tUNK UNK || UNK space\tREL\tmethod used for task\n", "0.369717\tUNK UNK || UNK space\tREL\tmethod used for task\n", "0.369717\tUNK space || UNK\tREL\tmethod used for task\n", "0.369717\tUNK UNK space || UNK\tREL\tmethod used for task\n", "0.369717\tUNK space || UNK\tREL\tmethod used for task\n", "0.369717\tUNK space || UNK\tREL\tmethod used for task\n", "0.369717\tUNK space || UNK\tREL\tmethod used for task\n", "0.369717\tUNK UNK || UNK space\tREL\tmethod used for task\n", "0.369717\tUNK UNK || UNK space\tREL\tmethod used for task\n", "0.369717\tUNK UNK || UNK UNK space\tREL\tmethod used for task\n", "0.369717\tUNK space || UNK\tREL\tmethod used for task\n", "0.369717\tUNK space || UNK\tREL\tmethod used for task\n", "0.369717\tUNK UNK || UNK UNK space\tREL\tmethod used for task\n", "0.369047\toptimal UNK UNK || test systems\tREL\tmethod used for task\n", "0.368401\tcomplex data || UNK\tREL\tmethod used for task\n", "0.368401\tcomplex data || UNK\tREL\tmethod used for task\n", "0.366761\tsensor data || data\tREL\tmethod used for task\n", "0.366290\tfrequency domain || UNK relations\tREL\tmethod used for task\n", "0.366171\tUNK model || solution technique\tREL\tmethod used for task\n", "0.366085\tnumerical solution || UNK UNK method\tREL\tmethod used for task\n", "0.364500\tsolution quality || optimization\tREL\tmethod used for task\n", "0.359378\tartificial neural network || UNK\tREL\tmethod used for task\n", "0.359378\tartificial neural network || UNK\tREL\tmethod used for task\n", "0.359378\tartificial neural network || UNK\tREL\tmethod used for task\n", "0.359378\tartificial neural network || UNK\tREL\tmethod used for task\n", "0.359378\tartificial neural network || UNK\tREL\tmethod used for task\n", "0.357520\tlinear systems || UNK\tREL\tmethod used for task\n", "0.356798\tperformance UNK || quality\tREL\tmethod used for task\n", "0.355923\tUNK networks || network size\tREL\tmethod used for task\n", "0.354484\tlinear UNK || computational\tREL\tmethod used for task\n", "0.351943\tcomputational cost || UNK performance\tREL\tmethod used for task\n", "0.351393\tcomputational study || UNK search\tREL\tmethod used for task\n", "0.351102\tUNK UNK || site\tREL\tmethod used for task\n", "0.351102\tUNK site || UNK\tREL\tmethod used for task\n", "0.351102\tUNK UNK || site\tREL\tmethod used for task\n", "0.350790\tUNK UNK || UNK UNK UNK planning\tREL\tmethod used for task\n", "0.350790\tUNK UNK || UNK planning\tREL\tmethod used for task\n", "0.350790\tUNK UNK || planning\tREL\tmethod used for task\n", "0.350790\tUNK UNK || planning\tREL\tmethod used for task\n", "0.350790\tUNK planning || UNK\tREL\tmethod used for task\n", "0.350790\tUNK planning || UNK\tREL\tmethod used for task\n", "0.350790\tUNK planning || UNK\tREL\tmethod used for task\n", "0.350790\tUNK planning || UNK\tREL\tmethod used for task\n", "0.350790\tUNK planning || UNK\tREL\tmethod used for task\n", "0.350790\tUNK planning || UNK\tREL\tmethod used for task\n", "0.345403\tpredictive UNK || UNK\tREL\tmethod used for task\n", "0.341980\tcomputational UNK || regression models\tREL\tmethod used for task\n", "0.339855\tUNK UNK UNK || surface\tREL\tmethod used for task\n", "0.339855\tUNK UNK || surface\tREL\tmethod used for task\n", "0.339855\tUNK UNK || UNK surface\tREL\tmethod used for task\n", "0.339855\tUNK UNK || surface\tREL\tmethod used for task\n", "0.339855\tUNK UNK || surface\tREL\tmethod used for task\n", "0.339855\tUNK UNK || surface\tREL\tmethod used for task\n", "0.339855\tUNK UNK || surface\tREL\tmethod used for task\n", "0.339855\tUNK UNK || surface\tREL\tmethod used for task\n", "0.339855\tUNK UNK || surface\tREL\tmethod used for task\n", "0.339855\tUNK UNK || surface\tREL\tmethod used for task\n", "0.339855\tUNK UNK || surface\tREL\tmethod used for task\n", "0.339855\tsurface UNK || UNK\tREL\tmethod used for task\n", "0.339855\tUNK surface || UNK\tREL\tmethod used for task\n", "0.339855\tsurface UNK || UNK\tREL\tmethod used for task\n", "0.339855\tsurface UNK || UNK\tREL\tmethod used for task\n", "0.339855\tsurface UNK || UNK\tREL\tmethod used for task\n", "0.339855\tsurface UNK || UNK\tREL\tmethod used for task\n", "0.339236\tUNK UNK || network\tREL\tmethod used for task\n", "0.339236\tUNK UNK || network\tREL\tmethod used for task\n", "0.339236\tUNK UNK || network\tREL\tmethod used for task\n", "0.339236\tUNK UNK || network\tREL\tmethod used for task\n", "0.339236\tUNK UNK UNK || UNK network\tREL\tmethod used for task\n", "0.339236\tnetwork UNK || UNK\tREL\tmethod used for task\n", "0.339236\tnetwork UNK || UNK\tREL\tmethod used for task\n", "0.339236\tUNK network || UNK\tREL\tmethod used for task\n", "0.339236\tUNK network || UNK\tREL\tmethod used for task\n", "0.339236\tnetwork UNK || UNK\tREL\tmethod used for task\n", "0.339236\tnetwork UNK || UNK\tREL\tmethod used for task\n", "0.339236\tnetwork UNK || UNK\tREL\tmethod used for task\n", "0.339236\tnetwork UNK || UNK\tREL\tmethod used for task\n", "0.339236\tnetwork UNK || UNK\tREL\tmethod used for task\n", "0.339236\tUNK network || UNK\tREL\tmethod used for task\n", "0.339236\tUNK UNK || UNK network\tREL\tmethod used for task\n", "0.339236\tUNK UNK || UNK network\tREL\tmethod used for task\n", "0.339236\tUNK UNK || UNK network\tREL\tmethod used for task\n", "0.339236\tUNK UNK || UNK network\tREL\tmethod used for task\n", "0.338227\tUNK UNK || UNK variables\tREL\tmethod used for task\n", "0.338227\tUNK UNK || UNK variables\tREL\tmethod used for task\n", "0.338227\tUNK UNK || UNK variables\tREL\tmethod used for task\n", "0.338227\tUNK UNK || UNK variables\tREL\tmethod used for task\n", "0.338227\tUNK variables || UNK\tREL\tmethod used for task\n", "0.338227\tUNK UNK || UNK variables\tREL\tmethod used for task\n", "0.338227\tUNK variables || UNK\tREL\tmethod used for task\n", "0.338227\tUNK UNK || UNK variables\tREL\tmethod used for task\n", "0.338227\tUNK variables || UNK\tREL\tmethod used for task\n", "0.338227\tUNK UNK || UNK variables\tREL\tmethod used for task\n", "0.338013\tUNK parameters || computational\tREL\tmethod used for task\n", "0.337662\tcase study || UNK models\tREL\tmethod used for task\n", "0.337662\tcase study || UNK models\tREL\tmethod used for task\n", "0.337216\tUNK data || predictive\tREL\tmethod used for task\n", "0.334811\tcase study || optimal\tREL\tmethod used for task\n", "0.334728\tsimulation model || wind\tREL\tmethod used for task\n", "0.333772\tautomatic detection || UNK\tREL\tmethod used for task\n", "0.333772\tautomatic detection || UNK\tREL\tmethod used for task\n", "0.333772\tautomatic detection || UNK\tREL\tmethod used for task\n", "0.333772\tautomatic detection || UNK\tREL\tmethod used for task\n", "0.333735\tshape space || UNK\tREL\tmethod used for task\n", "0.333720\tUNK UNK || UNK solution\tREL\tmethod used for task\n", "0.333720\tUNK UNK || UNK solution\tREL\tmethod used for task\n", "0.333720\tsolution UNK || UNK\tREL\tmethod used for task\n", "0.333720\tsolution UNK || UNK\tREL\tmethod used for task\n", "0.333720\tUNK solution || UNK\tREL\tmethod used for task\n", "0.333720\tUNK solution || UNK\tREL\tmethod used for task\n", "0.333720\tUNK UNK || solution\tREL\tmethod used for task\n", "0.333720\tUNK UNK || solution\tREL\tmethod used for task\n", "0.333720\tUNK UNK || solution\tREL\tmethod used for task\n", "0.333720\tUNK UNK || solution\tREL\tmethod used for task\n", "0.333720\tUNK UNK || solution\tREL\tmethod used for task\n", "0.333720\tUNK UNK || solution\tREL\tmethod used for task\n", "0.333720\tUNK UNK || solution\tREL\tmethod used for task\n", "0.333720\tUNK UNK || solution\tREL\tmethod used for task\n", "0.333720\tUNK UNK || solution\tREL\tmethod used for task\n", "0.333382\tcase study || UNK model\tREL\tmethod used for task\n", "0.333382\tcase study || UNK model\tREL\tmethod used for task\n", "0.333382\tcase study || UNK model\tREL\tmethod used for task\n", "0.331557\tUNK UNK || UNK linear regression models\tREL\tmethod used for task\n", "0.331557\tUNK UNK || UNK linear regression models\tREL\tmethod used for task\n", "0.331121\tUNK network || UNK data\tREL\tmethod used for task\n", "0.331121\tUNK network || UNK data\tREL\tmethod used for task\n", "0.331121\tUNK network || UNK data\tREL\tmethod used for task\n", "0.330124\tUNK variables || UNK data\tREL\tmethod used for task\n", "0.329766\tUNK UNK || frequency\tREL\tmethod used for task\n", "0.329766\tfrequency UNK || UNK\tREL\tmethod used for task\n", "0.329766\tUNK frequency UNK || UNK\tREL\tmethod used for task\n", "0.329766\tUNK frequency UNK || UNK\tREL\tmethod used for task\n", "0.326526\tUNK UNK UNK search || solution space\tREL\tmethod used for task\n", "0.323856\tcomputational UNK || UNK analysis\tREL\tmethod used for task\n", "0.319607\tcost UNK || optimal solution\tREL\tmethod used for task\n", "0.313721\tUNK analysis || linear\tREL\tmethod used for task\n", "0.313721\tUNK analysis || linear\tREL\tmethod used for task\n", "0.313721\tlinear UNK analysis || UNK\tREL\tmethod used for task\n", "0.311525\tUNK size || UNK\tREL\tmethod used for task\n", "0.311525\tUNK size || UNK\tREL\tmethod used for task\n", "0.311525\tUNK UNK || UNK size\tREL\tmethod used for task\n", "0.311525\tUNK UNK || UNK size\tREL\tmethod used for task\n", "0.311525\tUNK UNK || size\tREL\tmethod used for task\n", "0.304312\tsimulation study || UNK search\tREL\tmethod used for task\n", "0.303768\tUNK size || UNK data\tREL\tmethod used for task\n", "0.296881\tmodel parameters || cost parameters\tREL\tmethod used for task\n", "0.294333\tUNK model || wind turbine\tREL\tmethod used for task\n", "0.293296\twind UNK || UNK\tREL\tmethod used for task\n", "0.293296\tUNK UNK || wind\tREL\tmethod used for task\n", "0.293296\twind UNK || UNK\tREL\tmethod used for task\n", "0.293296\tUNK UNK UNK || UNK wind\tREL\tmethod used for task\n", "0.286947\tUNK study || computational modeling\tREL\tmethod used for task\n", "0.285130\tUNK UNK || product development\tREL\tmethod used for task\n", "0.281748\tnumerical analysis || UNK\tREL\tmethod used for task\n", "0.281748\tUNK UNK || numerical analysis\tREL\tmethod used for task\n", "0.281748\tnumerical UNK || UNK analysis\tREL\tmethod used for task\n", "0.280576\tautomatic UNK || UNK\tREL\tmethod used for task\n", "0.280576\tautomatic UNK || UNK\tREL\tmethod used for task\n", "0.280576\tautomatic UNK || UNK\tREL\tmethod used for task\n", "0.280576\tautomatic UNK || UNK\tREL\tmethod used for task\n", "0.280576\tautomatic UNK || UNK\tREL\tmethod used for task\n", "0.280576\tautomatic UNK || UNK\tREL\tmethod used for task\n", "0.280576\tautomatic UNK || UNK\tREL\tmethod used for task\n", "0.280576\tautomatic UNK || UNK\tREL\tmethod used for task\n", "0.280576\tUNK UNK || automatic\tREL\tmethod used for task\n", "0.278878\tsimulation UNK || analysis\tREL\tmethod used for task\n", "0.278694\tproblem UNK || solution quality\tREL\tmethod used for task\n", "0.277765\tUNK data || product development\tREL\tmethod used for task\n", "0.276388\ttest UNK || UNK\tREL\tmethod used for task\n", "0.276388\ttest UNK || UNK\tREL\tmethod used for task\n", "0.276388\ttest UNK || UNK\tREL\tmethod used for task\n", "0.276388\ttest UNK || UNK\tREL\tmethod used for task\n", "0.276388\tUNK UNK || UNK test\tREL\tmethod used for task\n", "0.276388\tUNK UNK || test\tREL\tmethod used for task\n", "0.276388\tUNK UNK || test\tREL\tmethod used for task\n", "0.276388\tUNK UNK || test\tREL\tmethod used for task\n", "0.276388\tUNK UNK || test\tREL\tmethod used for task\n", "0.276388\tUNK UNK || test\tREL\tmethod used for task\n", "0.276388\tUNK UNK || UNK test\tREL\tmethod used for task\n", "0.276388\tUNK UNK || UNK test\tREL\tmethod used for task\n", "0.275610\tUNK UNK || numerical simulation\tREL\tmethod used for task\n", "0.275610\tnumerical simulation || UNK\tREL\tmethod used for task\n", "0.275116\tdetection UNK || UNK study\tREL\tmethod used for task\n", "0.267260\tUNK UNK || UNK quality\tREL\tmethod used for task\n", "0.267260\tUNK quality || UNK\tREL\tmethod used for task\n", "0.267260\tUNK quality || UNK\tREL\tmethod used for task\n", "0.267260\tUNK quality || UNK\tREL\tmethod used for task\n", "0.267260\tUNK quality || UNK\tREL\tmethod used for task\n", "0.267260\tquality UNK || UNK\tREL\tmethod used for task\n", "0.267260\tUNK quality || UNK\tREL\tmethod used for task\n", "0.267260\tUNK quality || UNK\tREL\tmethod used for task\n", "0.267260\tUNK UNK || UNK quality\tREL\tmethod used for task\n", "0.267260\tUNK UNK || UNK quality\tREL\tmethod used for task\n", "0.267260\tUNK UNK || UNK quality\tREL\tmethod used for task\n", "0.267260\tUNK quality UNK || UNK\tREL\tmethod used for task\n", "0.267260\tquality UNK || UNK\tREL\tmethod used for task\n", "0.267260\tquality UNK || UNK\tREL\tmethod used for task\n", "0.267260\tquality UNK || UNK\tREL\tmethod used for task\n", "0.267260\tUNK quality || UNK\tREL\tmethod used for task\n", "0.267260\tUNK quality || UNK\tREL\tmethod used for task\n", "0.267260\tUNK UNK || UNK quality\tREL\tmethod used for task\n", "0.267260\tquality UNK || UNK\tREL\tmethod used for task\n", "0.267260\tquality UNK || UNK\tREL\tmethod used for task\n", "0.267260\tUNK UNK || quality\tREL\tmethod used for task\n", "0.267260\tUNK quality || UNK\tREL\tmethod used for task\n", "0.265315\tUNK UNK || boundary\tREL\tmethod used for task\n", "0.265315\tboundary UNK || UNK\tREL\tmethod used for task\n", "0.264850\tspatial UNK || UNK\tREL\tmethod used for task\n", "0.264850\tUNK UNK || UNK spatial\tREL\tmethod used for task\n", "0.264850\tspatial UNK || UNK\tREL\tmethod used for task\n", "0.264850\tspatial UNK || UNK\tREL\tmethod used for task\n", "0.264850\tspatial UNK || UNK\tREL\tmethod used for task\n", "0.264850\tspatial UNK || UNK\tREL\tmethod used for task\n", "0.264850\tspatial UNK || UNK\tREL\tmethod used for task\n", "0.264850\tspatial UNK || UNK\tREL\tmethod used for task\n", "0.264850\tUNK UNK || spatial\tREL\tmethod used for task\n", "0.264850\tUNK UNK UNK || UNK spatial\tREL\tmethod used for task\n", "0.263724\tsimulation study || UNK model\tREL\tmethod used for task\n", "0.263125\tcomputational cost || UNK\tREL\tmethod used for task\n", "0.263125\tcomputational cost || UNK\tREL\tmethod used for task\n", "0.263125\tcomputational cost || UNK\tREL\tmethod used for task\n", "0.263125\tcomputational cost || UNK\tREL\tmethod used for task\n", "0.263125\tcomputational cost || UNK\tREL\tmethod used for task\n", "0.263125\tUNK UNK || computational cost\tREL\tmethod used for task\n", "0.263083\tUNK neural network || UNK systems\tREL\tmethod used for task\n", "0.260189\tUNK quality || data\tREL\tmethod used for task\n", "0.260189\tdata quality || UNK\tREL\tmethod used for task\n", "0.258276\tboundary UNK || data\tREL\tmethod used for task\n", "0.257461\tinput space || inverse\tREL\tmethod used for task\n", "0.256125\tcomputational cost || UNK data\tREL\tmethod used for task\n", "0.253477\tUNK UNK || UNK UNK regression\tREL\tmethod used for task\n", "0.253477\tUNK UNK || UNK regression\tREL\tmethod used for task\n", "0.253477\tUNK UNK || UNK regression\tREL\tmethod used for task\n", "0.253477\tUNK UNK || UNK regression\tREL\tmethod used for task\n", "0.253477\tUNK UNK UNK || regression\tREL\tmethod used for task\n", "0.253477\tUNK UNK UNK || UNK UNK regression\tREL\tmethod used for task\n", "0.253477\tUNK UNK || regression\tREL\tmethod used for task\n", "0.253477\tUNK regression || UNK\tREL\tmethod used for task\n", "0.253477\tUNK UNK regression || UNK\tREL\tmethod used for task\n", "0.253477\tregression UNK || UNK\tREL\tmethod used for task\n", "0.253477\tUNK regression || UNK\tREL\tmethod used for task\n", "0.253477\tUNK regression || UNK\tREL\tmethod used for task\n", "0.253477\tUNK UNK regression || UNK\tREL\tmethod used for task\n", "0.253240\tdata UNK || data quality\tREL\tmethod used for task\n", "0.253240\tdata quality || UNK data\tREL\tmethod used for task\n", "0.245468\tengineering UNK || UNK UNK UNK systems\tREL\tmethod used for task\n", "0.245468\tUNK UNK || systems engineering\tREL\tmethod used for task\n", "0.241273\tUNK UNK || input\tREL\tmethod used for task\n", "0.241273\tUNK UNK || input\tREL\tmethod used for task\n", "0.241273\tUNK UNK || UNK input\tREL\tmethod used for task\n", "0.241273\tUNK input || UNK\tREL\tmethod used for task\n", "0.241273\tUNK UNK || input\tREL\tmethod used for task\n", "0.240934\tUNK UNK || complex systems\tREL\tmethod used for task\n", "0.240934\tcomplex systems || UNK\tREL\tmethod used for task\n", "0.238495\tUNK UNK || UNK sensitivity analysis\tREL\tmethod used for task\n", "0.238495\tsensitivity analysis || UNK\tREL\tmethod used for task\n", "0.238495\tsensitivity analysis || UNK\tREL\tmethod used for task\n", "0.238495\tsensitivity analysis || UNK\tREL\tmethod used for task\n", "0.238495\tsensitivity analysis || UNK\tREL\tmethod used for task\n", "0.238495\tsensitivity analysis || UNK\tREL\tmethod used for task\n", "0.238495\tsensitivity analysis || UNK\tREL\tmethod used for task\n", "0.238495\tsensitivity analysis || UNK\tREL\tmethod used for task\n", "0.238495\tsensitivity analysis || UNK\tREL\tmethod used for task\n", "0.238068\tUNK algorithm || solution quality\tREL\tmethod used for task\n", "0.238068\tUNK algorithm || solution quality\tREL\tmethod used for task\n", "0.238068\tUNK UNK algorithm || solution quality\tREL\tmethod used for task\n", "0.238068\tUNK UNK algorithm || solution quality\tREL\tmethod used for task\n", "0.231640\tsolution quality || search\tREL\tmethod used for task\n", "0.229404\tproduct UNK || product\tREL\tmethod used for task\n", "0.229139\tcost analysis || UNK\tREL\tmethod used for task\n", "0.228069\thistorical data || UNK\tREL\tmethod used for task\n", "0.228068\tUNK study || UNK\tREL\tmethod used for task\n", "0.228068\tUNK study || UNK\tREL\tmethod used for task\n", "0.228068\tUNK study || UNK\tREL\tmethod used for task\n", "0.228068\tUNK study || UNK\tREL\tmethod used for task\n", "0.228068\tUNK study || UNK\tREL\tmethod used for task\n", "0.228068\tUNK study || UNK\tREL\tmethod used for task\n", "0.228068\tUNK study || UNK\tREL\tmethod used for task\n", "0.228068\tUNK study || UNK\tREL\tmethod used for task\n", "0.221720\tUNK study || UNK UNK data\tREL\tmethod used for task\n", "0.221720\tUNK data || UNK study\tREL\tmethod used for task\n", "0.216835\tpredictive UNK || UNK systems\tREL\tmethod used for task\n", "0.212218\tnetwork UNK || UNK systems\tREL\tmethod used for task\n", "0.211429\tUNK study || experimental analysis\tREL\tmethod used for task\n", "0.211429\tUNK study || experimental analysis\tREL\tmethod used for task\n", "0.211429\texperimental study || UNK analysis\tREL\tmethod used for task\n", "0.205943\tcomputational UNK || solution\tREL\tmethod used for task\n", "0.205943\tcomputational UNK || solution\tREL\tmethod used for task\n", "0.205943\tcomputational UNK || UNK solution\tREL\tmethod used for task\n", "0.205193\tfrequency UNK || UNK systems\tREL\tmethod used for task\n", "0.203317\tlinear regression UNK || UNK performance\tREL\tmethod used for task\n", "0.197986\tUNK model || solution quality\tREL\tmethod used for task\n", "0.197986\tUNK model || solution quality\tREL\tmethod used for task\n", "0.196944\tUNK space || UNK UNK simulation\tREL\tmethod used for task\n", "0.191870\tUNK UNK || size systems\tREL\tmethod used for task\n", "0.191309\tUNK parameters || surface\tREL\tmethod used for task\n", "0.190436\tUNK space representation || UNK\tREL\tmethod used for task\n", "0.190436\tUNK space representation || UNK\tREL\tmethod used for task\n", "0.190436\tUNK space representation || UNK\tREL\tmethod used for task\n", "0.184283\tlinear UNK || linear UNK analysis\tREL\tmethod used for task\n", "0.176714\tUNK UNK network || simulation\tREL\tmethod used for task\n", "0.175194\tnumerical solution || UNK\tREL\tmethod used for task\n", "0.175194\tnumerical solution || UNK\tREL\tmethod used for task\n", "0.175194\tnumerical solution || UNK\tREL\tmethod used for task\n", "0.169875\tautomatic UNK || UNK UNK systems\tREL\tmethod used for task\n", "0.160641\tsystems UNK || UNK quality\tREL\tmethod used for task\n", "0.158863\tcomputational UNK || UNK quality\tREL\tmethod used for task\n", "0.158152\tperformance UNK || simulation study\tREL\tmethod used for task\n", "0.153142\tsensor UNK || sensor\tREL\tmethod used for task\n", "0.153142\tsensor UNK || sensor\tREL\tmethod used for task\n", "0.151779\tUNK regression model || automatic\tREL\tmethod used for task\n", "0.151747\tUNK analysis || wind\tREL\tmethod used for task\n", "0.147097\tUNK UNK UNK UNK || case study\tREL\tmethod used for task\n", "0.147097\tcase study || UNK\tREL\tmethod used for task\n", "0.147097\tcase study || UNK\tREL\tmethod used for task\n", "0.147097\tcase study || UNK\tREL\tmethod used for task\n", "0.147097\tcase study || UNK\tREL\tmethod used for task\n", "0.147097\tcase study || UNK\tREL\tmethod used for task\n", "0.147097\tcase study || UNK\tREL\tmethod used for task\n", "0.147097\tUNK UNK || case study\tREL\tmethod used for task\n", "0.147097\tcase study || UNK\tREL\tmethod used for task\n", "0.147097\tUNK UNK || case study\tREL\tmethod used for task\n", "0.147097\tcase study || UNK\tREL\tmethod used for task\n", "0.147097\tcase study || UNK\tREL\tmethod used for task\n", "0.147097\tcase study || UNK\tREL\tmethod used for task\n", "0.147097\tcase study || UNK\tREL\tmethod used for task\n", "0.144631\tUNK analysis || numerical analysis\tREL\tmethod used for task\n", "0.143915\tautomatic analysis || UNK\tREL\tmethod used for task\n", "0.143690\tUNK UNK UNK || linear regression\tREL\tmethod used for task\n", "0.143690\tUNK UNK UNK || linear regression\tREL\tmethod used for task\n", "0.143690\tlinear regression || UNK\tREL\tmethod used for task\n", "0.143690\tlinear regression || UNK\tREL\tmethod used for task\n", "0.143690\tlinear regression || UNK\tREL\tmethod used for task\n", "0.143690\tUNK UNK || linear regression\tREL\tmethod used for task\n", "0.143662\tUNK UNK || heterogeneous network\tREL\tmethod used for task\n", "0.140322\tUNK UNK UNK neural network || surface\tREL\tmethod used for task\n", "0.139989\tnetwork UNK || UNK neural network\tREL\tmethod used for task\n", "0.139694\tUNK UNK || surface meshes\tREL\tmethod used for task\n", "0.139267\tUNK data || linear regression\tREL\tmethod used for task\n", "0.138641\tcost UNK || solution\tREL\tmethod used for task\n", "0.137700\tUNK UNK || simulation test\tREL\tmethod used for task\n", "0.137700\tUNK UNK || simulation test\tREL\tmethod used for task\n", "0.137044\tregression models || UNK regression\tREL\tmethod used for task\n", "0.135330\tcase study || experimental analysis\tREL\tmethod used for task\n", "0.134418\tUNK analysis || spatial\tREL\tmethod used for task\n", "0.132806\tnumerical UNK || boundary\tREL\tmethod used for task\n", "0.132688\tcomputational study || UNK\tREL\tmethod used for task\n", "0.132688\tcomputational study || UNK\tREL\tmethod used for task\n", "0.132688\tcomputational study || UNK\tREL\tmethod used for task\n", "0.128724\tcase study || solution algorithm\tREL\tmethod used for task\n", "0.128552\tcomputational study || UNK data\tREL\tmethod used for task\n", "0.127673\tUNK UNK || regression analysis\tREL\tmethod used for task\n", "0.127673\tregression analysis || UNK\tREL\tmethod used for task\n", "0.127494\tUNK UNK || input parameters\tREL\tmethod used for task\n", "0.126948\tUNK cost || UNK size\tREL\tmethod used for task\n", "0.125809\tsensitivity analysis || UNK parameters\tREL\tmethod used for task\n", "0.120424\tUNK UNK || solution space\tREL\tmethod used for task\n", "0.119535\tUNK parameters || UNK study\tREL\tmethod used for task\n", "0.114633\tUNK performance || solution quality\tREL\tmethod used for task\n", "0.112967\thistorical data || UNK analysis\tREL\tmethod used for task\n", "0.107271\tUNK surface UNK || solution\tREL\tmethod used for task\n", "0.107007\tUNK solution || UNK network\tREL\tmethod used for task\n", "0.106577\tUNK solution || UNK variables\tREL\tmethod used for task\n", "0.106429\tsimulation study || data\tREL\tmethod used for task\n", "0.103656\tneural network || UNK quality\tREL\tmethod used for task\n", "0.097923\tUNK size || UNK surface\tREL\tmethod used for task\n", "0.097191\tUNK UNK || regression neural network\tREL\tmethod used for task\n", "0.097191\tUNK UNK UNK || regression neural network\tREL\tmethod used for task\n", "0.091666\tmodel parameters || input variables\tREL\tmethod used for task\n", "0.091589\tneural network || input\tREL\tmethod used for task\n", "0.089953\twind UNK || UNK variables\tREL\tmethod used for task\n", "0.087100\tUNK size || UNK size\tREL\tmethod used for task\n", "0.083431\tnumerical simulation || UNK network\tREL\tmethod used for task\n", "0.082986\tUNK systems || case study\tREL\tmethod used for task\n", "0.078453\tquality solution || UNK\tREL\tmethod used for task\n", "0.078453\tUNK UNK || solution quality\tREL\tmethod used for task\n", "0.078453\tUNK UNK || solution quality\tREL\tmethod used for task\n", "0.078453\tUNK UNK || solution quality\tREL\tmethod used for task\n", "0.078453\tUNK UNK || solution quality\tREL\tmethod used for task\n", "0.078453\tUNK UNK || solution quality\tREL\tmethod used for task\n", "0.078453\tUNK UNK || solution quality\tREL\tmethod used for task\n", "0.078453\tUNK UNK || solution quality\tREL\tmethod used for task\n", "0.078453\tUNK UNK || solution quality\tREL\tmethod used for task\n", "0.078453\tUNK UNK || solution quality\tREL\tmethod used for task\n", "0.077361\tsimulation study || supply network\tREL\tmethod used for task\n", "0.076299\tspatial frequency || UNK\tREL\tmethod used for task\n", "0.075294\tpopulation size || UNK\tREL\tmethod used for task\n", "0.074535\ttest UNK || UNK size\tREL\tmethod used for task\n", "0.070700\tUNK network || input\tREL\tmethod used for task\n", "0.070405\tUNK UNK || input variables\tREL\tmethod used for task\n", "0.070405\tUNK UNK || input variables\tREL\tmethod used for task\n", "0.070405\tUNK UNK || input variables\tREL\tmethod used for task\n", "0.068154\tnumerical UNK || case study\tREL\tmethod used for task\n", "0.067452\tlinear regression || UNK analysis\tREL\tmethod used for task\n", "0.067452\tUNK analysis || linear regression\tREL\tmethod used for task\n", "0.067452\tUNK analysis || linear regression\tREL\tmethod used for task\n", "0.064511\tUNK study || UNK solution\tREL\tmethod used for task\n", "0.062585\tsimulation test || simulation\tREL\tmethod used for task\n", "0.062585\tsimulation test || simulation\tREL\tmethod used for task\n", "0.062585\tsimulation test || simulation\tREL\tmethod used for task\n", "0.062585\tsimulation UNK || simulation test\tREL\tmethod used for task\n", "0.060170\tsolution technique || cost\tREL\tmethod used for task\n", "0.058376\tquality UNK || quality\tREL\tmethod used for task\n", "0.052801\tsensitivity analysis || test\tREL\tmethod used for task\n", "0.051240\tnumerical analysis || UNK study\tREL\tmethod used for task\n", "0.050985\tregression UNK || UNK UNK regression\tREL\tmethod used for task\n", "0.050985\tregression UNK || UNK regression\tREL\tmethod used for task\n", "0.049960\tUNK study || test\tREL\tmethod used for task\n", "0.042221\tcomputational UNK || quality solution\tREL\tmethod used for task\n", "0.042221\tcomputational UNK || solution quality\tREL\tmethod used for task\n", "0.038221\tUNK systems || input variables\tREL\tmethod used for task\n", "0.037647\tUNK parameters || solution quality\tREL\tmethod used for task\n", "0.035397\tsolution quality || UNK analysis\tREL\tmethod used for task\n", "0.034477\tcomputational study || UNK solution\tREL\tmethod used for task\n", "0.034477\tcomputational study || solution\tREL\tmethod used for task\n", "0.034369\tsimulation UNK || solution quality\tREL\tmethod used for task\n", "0.025864\tregression UNK || linear regression\tREL\tmethod used for task\n", "0.025864\tregression UNK || linear regression\tREL\tmethod used for task\n", "0.025864\tlinear regression || UNK UNK regression\tREL\tmethod used for task\n", "0.025864\tlinear regression || UNK UNK regression\tREL\tmethod used for task\n", "0.024553\tsensitivity analysis || case study\tREL\tmethod used for task\n", "0.023194\tcase study || UNK study\tREL\tmethod used for task\n", "0.016748\tregression UNK || regression neural network\tREL\tmethod used for task\n", "0.016748\tUNK UNK regression || regression neural network\tREL\tmethod used for task\n", "0.015759\tcase study || solution technique\tREL\tmethod used for task\n", "0.013368\tpopulation size || computational cost\tREL\tmethod used for task\n", "0.008348\tlinear regression || regression neural network\tREL\tmethod used for task\n", "0.008348\tlinear regression || regression neural network\tREL\tmethod used for task\n", "0.004167\tspatial relations || space layouts\tREL\tmethod used for task\n" ] } ], "source": [ "# show predictions\n", "ents_test = [ie.reverse_dict_lookup(dictionary_ents_rev, e) for e in ents_test_pos]\n", "rels_test = [ie.reverse_dict_lookup(dictionary_rels_rev, r) for r in rels_test_pos]\n", "testresults = sorted(zip(test_scores, ents_test, rels_test), key=lambda t: t[0], reverse=True) # sort for decreasing score\n", "\n", "print(\"\\nTest predictions by decreasing probability:\")\n", "for score, tup, rel in testresults:\n", " print('%f\\t%s\\tREL\\t%s' % (score, \" \".join(tup), \" \".join(rel)))" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "Test prediction probabilities are obtained by scoring each test instances with:\n", "\n", "$\\mathcal{ \\sigma ( v_{e} * a_{r} )}$\n", "\n", "* Note that as input for the latent feature representation, we discarded words that only appeared twice\n", " * Hence, for those words we did not learn a representation, denoted here by 'UNK'\n", "* This is also typically done for other feature representations, as if we only see a feature once, it is difficult to learn weights for it" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "**Thought Exercises**: \n", "* The scores shown here are for the relation 'method used for task'. However, we could also use our model to score the compatibility of entity pairs with other relations, e.g. 'demonstrates XXXXX for XXXXXX'. How could this be done here?\n", "* How could we get around the problem of unseen words, as described above?\n", "* What other possible problems can you see with the above formulation of universal schema relation extraction?\n", "* What possible problems can you see with using latent word representations?" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "## Background Material\n", "\n", "* Jurafky, Dan and Martin, James H. (2016). Speech and Language Processing, Chapter 21 (Information Extraction): https://web.stanford.edu/~jurafsky/slp3/21.pdf\n", "\n", "* Riedel, Sebastian and Yao, Limin and McCallum, Andrew and Marlin, Benjamin M. (2013). Extraction with Matrix Factorization and Universal Schemas. Proceedings of NAACL. http://www.aclweb.org/anthology/N13-1008" ] } ], "metadata": { "celltoolbar": "Slideshow", "kernelspec": { "display_name": "Python 3", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.6.2" } }, "nbformat": 4, "nbformat_minor": 1 }