{
 "cells": [
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "96ec678e-b20c-4213-8616-542010f46342",
   "metadata": {},
   "source": [
    "# Dirty ER\n",
    "\n",
    "---\n",
    "\n",
    "In this notebook we present the pyJedAI approach in the well-known ABT-BUY dataset. Dirty ER, is the process of dedeplication of one set."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5274855c-ba95-49b1-ba68-4f50ca2bbd89",
   "metadata": {},
   "source": [
    "# Instalation\n",
    "\n",
    "pyJedAI is an open-source library that can be installed from PyPI.\n",
    "\n",
    "For more: [pypi.org/project/pyjedai/](https://pypi.org/project/pyjedai/)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "4697d149-c1a4-4767-9ed1-14444485e409",
   "metadata": {},
   "outputs": [],
   "source": [
    "!python --version"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "776843a2-570d-4d87-bb1d-b5b61f07b1da",
   "metadata": {},
   "outputs": [],
   "source": [
    "!pip install pyjedai -U"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "6d2e5cf7-ff2e-4271-9242-fe3d638263e9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Name: pyjedai\n",
      "Version: 0.0.5\n",
      "Summary: An open-source library that builds powerful end-to-end Entity Resolution workflows.\n",
      "Home-page: \n",
      "Author: \n",
      "Author-email: Konstantinos Nikoletos <nikoletos.kon@gmail.com>, George Papadakis <gpapadis84@gmail.com>\n",
      "License: Apache Software License 2.0\n",
      "Location: c:\\users\\nikol\\anaconda3\\lib\\site-packages\n",
      "Requires: strsim, seaborn, matplotlib, optuna, networkx, faiss-cpu, scipy, rdflib, strsimpy, rdfpandas, regex, sentence-transformers, tqdm, nltk, pandas, pandocfilters, matplotlib-inline, numpy, pandas-profiling, PyYAML, transformers, tomli, gensim\n",
      "Required-by: \n"
     ]
    }
   ],
   "source": [
    "!pip show pyjedai"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "15d28272-269a-4e87-bb03-0a45a5492a06",
   "metadata": {},
   "source": [
    "Imports"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "a0890ce6-3a10-4e66-913f-78095bd786a1",
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import sys\n",
    "import pandas as pd\n",
    "import networkx\n",
    "from networkx import draw, Graph\n",
    "\n",
    "from pyjedai.utils import print_clusters, print_blocks, print_candidate_pairs\n",
    "from pyjedai.evaluation import Evaluation"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "af77914f-5e76-4da8-a0ad-1c53e0111a0f",
   "metadata": {
    "tags": []
   },
   "source": [
    "# Reading the dataset - Dirty ER example\n",
    "\n",
    "pyJedAI in order to perfrom needs only the tranformation of the initial data into a pandas DataFrame. Hence, pyJedAI can function in every structured or semi-structured data. In this case Abt-Buy dataset is provided as .csv files. \n",
    "\n",
    "<div align=\"center\">\n",
    "    <img align=\"center\" src=\"https://github.com/AI-team-UoA/pyJedAI/blob/main/documentation/reading-process.png?raw=true\" width=\"800\" />\n",
    "</div>\n",
    "\n",
    "\n",
    "## pyjedai <Data\\> module\n",
    "\n",
    "Data module offers a numpber of options\n",
    "- Selecting the parameters (columns) of the dataframe, in D1 (and in D2)\n",
    "- Prints a detailed text analysis\n",
    "- Stores a hidden mapping of the ids, and creates it if not exists."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "3d3feb89-1406-4c90-a1aa-dc2cf4707739",
   "metadata": {},
   "outputs": [],
   "source": [
    "from pyjedai.datamodel import Data\n",
    "\n",
    "d1 = pd.read_csv(\"./../data/der/cora/cora.csv\", sep='|')\n",
    "gt = pd.read_csv(\"./../data/der/cora/cora_gt.csv\", sep='|', header=None)\n",
    "attr = ['Entity Id','author', 'title']"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fda32323-c74d-4374-b322-5c11a175c3ea",
   "metadata": {},
   "source": [
    "Data is the connecting module of all steps of the workflow"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "e257597d-ea77-4090-ba34-e1038d8f9a0d",
   "metadata": {},
   "outputs": [],
   "source": [
    "data = Data(\n",
    "    dataset_1=d1,\n",
    "    id_column_name_1='Entity Id',\n",
    "    ground_truth=gt,\n",
    "    attributes_1=attr\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "93464edd-b88a-40d9-aa4d-7fe1523db662",
   "metadata": {},
   "source": [
    "# Workflow with Block Cleaning Methods\n",
    "\n",
    "In this notebook we created the bellow architecture:\n",
    "\n",
    "![workflow1-cora.png](https://github.com/AI-team-UoA/pyJedAI/blob/main/documentation/workflow1-cora.png?raw=true)\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9c068252-4a69-405a-a320-c2875ec08ea5",
   "metadata": {},
   "source": [
    "## Block Building\n",
    "\n",
    "It clusters entities into overlapping blocks in a lazy manner that relies on unsupervised blocking keys: every token in an attribute value forms a key. Blocks are then extracted, possibly using a transformation, based on its equality or on its similarity with other keys.\n",
    "\n",
    "The following methods are currently supported:\n",
    "\n",
    "- Standard/Token Blocking\n",
    "- Sorted Neighborhood\n",
    "- Extended Sorted Neighborhood\n",
    "- Q-Grams Blocking\n",
    "- Extended Q-Grams Blocking\n",
    "- Suffix Arrays Blocking\n",
    "- Extended Suffix Arrays Blocking"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "9c1b6213-a218-40cf-bc72-801b77d28da9",
   "metadata": {},
   "outputs": [],
   "source": [
    "from pyjedai.block_building import (\n",
    "    StandardBlocking,\n",
    "    QGramsBlocking,\n",
    "    SuffixArraysBlocking,\n",
    "    ExtendedSuffixArraysBlocking,\n",
    "    ExtendedQGramsBlocking\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "7ee34038-1352-440e-8c34-98c5cf036523",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1b3d9aac0ca94663b04832c034e8c18a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Suffix Arrays Blocking:   0%|          | 0/1295 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "bb = SuffixArraysBlocking(suffix_length=2)\n",
    "blocks = bb.build_blocks(data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "b0ac846d-0f13-4b90-b4c8-688054ed7ffe",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "***************************************************************************************************************************\n",
      "                                         Μethod:  Suffix Arrays Blocking\n",
      "***************************************************************************************************************************\n",
      "Method name: Suffix Arrays Blocking\n",
      "Parameters: \n",
      "\tSuffix length: 2\n",
      "\tMaximum Block Size: 53\n",
      "Runtime: 0.1680 seconds\n",
      "───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────\n",
      "Performance:\n",
      "\tPrecision:      4.28% \n",
      "\tRecall:        75.77%\n",
      "\tF1-score:       8.10%\n",
      "───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────\n"
     ]
    }
   ],
   "source": [
    "_ = bb.evaluate(blocks)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "62b067d8-fbc4-4689-ba72-7ab5baf88a70",
   "metadata": {
    "tags": []
   },
   "source": [
    "## Block Cleaning\n",
    "\n",
    "___Optional step___\n",
    "\n",
    "Its goal is to clean a set of overlapping blocks from unnecessary comparisons, which can be either redundant (i.e., repeated comparisons that have already been executed in a previously examined block) or superfluous (i.e., comparisons that involve non-matching entities). Its methods operate on the coarse level of individual blocks or entities."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "9c2c0e42-485a-444e-9161-975f30d21a02",
   "metadata": {},
   "outputs": [],
   "source": [
    "from pyjedai.block_cleaning import BlockFiltering"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "bf5c20ac-b16a-484d-82b0-61ecb9e7f3ea",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "68ce99a10fdf438bb6703aa33d26e037",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Block Filtering:   0%|          | 0/3 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "bc = BlockFiltering(ratio=0.9)\n",
    "blocks = bc.process(blocks, data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "25fd0be0-91c3-4d0b-b596-c66dccba3c79",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "***************************************************************************************************************************\n",
      "                                         Μethod:  Block Filtering\n",
      "***************************************************************************************************************************\n",
      "Method name: Block Filtering\n",
      "Parameters: \n",
      "\tRatio: 0.9\n",
      "Runtime: 0.0930 seconds\n",
      "───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────\n",
      "Performance:\n",
      "\tPrecision:      4.90% \n",
      "\tRecall:        75.56%\n",
      "\tF1-score:       9.21%\n",
      "───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────\n"
     ]
    }
   ],
   "source": [
    "_ = bc.evaluate(blocks)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9cd12048-bd0c-4571-ba70-488d46afcdd6",
   "metadata": {},
   "source": [
    "## Comparison Cleaning\n",
    "\n",
    "___Optional step___\n",
    "\n",
    "Similar to Block Cleaning, this step aims to clean a set of blocks from both redundant and superfluous comparisons. Unlike Block Cleaning, its methods operate on the finer granularity of individual comparisons.\n",
    "\n",
    "The following methods are currently supported:\n",
    "\n",
    "- Comparison Propagation\n",
    "- Cardinality Edge Pruning (CEP)\n",
    "- Cardinality Node Pruning (CNP)\n",
    "- Weighed Edge Pruning (WEP)\n",
    "- Weighed Node Pruning (WNP)\n",
    "- Reciprocal Cardinality Node Pruning (ReCNP)\n",
    "- Reciprocal Weighed Node Pruning (ReWNP)\n",
    "- BLAST\n",
    "\n",
    "Most of these methods are Meta-blocking techniques. All methods are optional, but competive, in the sense that only one of them can part of an ER workflow. For more details on the functionality of these methods, see here. They can be combined with one of the following weighting schemes:\n",
    "\n",
    "- Aggregate Reciprocal Comparisons Scheme (ARCS)\n",
    "- Common Blocks Scheme (CBS)\n",
    "- Enhanced Common Blocks Scheme (ECBS)\n",
    "- Jaccard Scheme (JS)\n",
    "- Enhanced Jaccard Scheme (EJS)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "7248d75f-215f-4efd-97f4-a2c9b969b9fc",
   "metadata": {},
   "outputs": [],
   "source": [
    "from pyjedai.block_cleaning import BlockPurging"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "a16bf7a0-180a-494b-8a0b-6a94d9c60deb",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "36d831c625014c5ca38987f72b7aa360",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Block Purging:   0%|          | 0/3192 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "bp = BlockPurging(smoothing_factor=0.008)\n",
    "blocks = bp.process(blocks, data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "8d408a88-9386-4d40-80f5-6b089bc6ad4b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "***************************************************************************************************************************\n",
      "                                         Μethod:  Block Purging\n",
      "***************************************************************************************************************************\n",
      "Method name: Block Purging\n",
      "Parameters: \n",
      "\tSmoothing factor: 0.008\n",
      "\tMax Comparisons per Block: 6.0\n",
      "Runtime: 0.0480 seconds\n",
      "───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────\n",
      "Performance:\n",
      "\tPrecision:      9.87% \n",
      "\tRecall:         0.72%\n",
      "\tF1-score:       1.33%\n",
      "───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────\n"
     ]
    }
   ],
   "source": [
    "_ = bp.evaluate(blocks)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "85196f27-d072-435e-a9f3-2142a83eafad",
   "metadata": {},
   "source": [
    "### Meta Blocking"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "1f7d75f3-6bed-482d-a572-c3b4927236a5",
   "metadata": {},
   "outputs": [],
   "source": [
    "from pyjedai.comparison_cleaning import (\n",
    "    WeightedEdgePruning,\n",
    "    WeightedNodePruning,\n",
    "    CardinalityEdgePruning,\n",
    "    CardinalityNodePruning,\n",
    "    BLAST,\n",
    "    ReciprocalCardinalityNodePruning,\n",
    "    ComparisonPropagation\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "c92e0ca3-5591-4620-b3f4-012a23637416",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f24c697770c8409e95721ba534b0edcf",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Weighted Edge Pruning:   0%|          | 0/1295 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "mb = WeightedEdgePruning(weighting_scheme='CBS')\n",
    "blocks = mb.process(blocks, data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "f469e387-e135-4945-b97f-da14d391c6b1",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "***************************************************************************************************************************\n",
      "                                         Μethod:  Weighted Edge Pruning\n",
      "***************************************************************************************************************************\n",
      "Method name: Weighted Edge Pruning\n",
      "Parameters: \n",
      "\tNode centric: False\n",
      "\tWeighting scheme: CBS\n",
      "Runtime: 0.0457 seconds\n",
      "───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────\n",
      "Performance:\n",
      "\tPrecision:     75.49% \n",
      "\tRecall:         0.45%\n",
      "\tF1-score:       0.89%\n",
      "───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────\n"
     ]
    }
   ],
   "source": [
    "_ = mb.evaluate(blocks)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6aeff39a-b51b-4166-a55b-f8452ec258a7",
   "metadata": {},
   "source": [
    "## Entity Matching\n",
    "\n",
    "It compares pairs of entity profiles, associating every pair with a similarity in [0,1]. Its output comprises the similarity graph, i.e., an undirected, weighted graph where the nodes correspond to entities and the edges connect pairs of compared entities."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "f479d967-8bac-4870-99bd-68c01e75747b",
   "metadata": {},
   "outputs": [],
   "source": [
    "from pyjedai.matching import EntityMatching"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "bcb37bf3-1da1-4830-9659-41cf85a536a9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "020d068b0b2848fa80d4ff4e0fd41236",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Entity Matching (jaccard):   0%|          | 0/59 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "attr = {\n",
    "    'author' : 0.6, \n",
    "    'title' : 0.4\n",
    "}\n",
    "\n",
    "EM = EntityMatching(\n",
    "    metric='jaccard',\n",
    "    similarity_threshold=0.5\n",
    ")\n",
    "\n",
    "pairs_graph = EM.predict(blocks, data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "4d606bfc-3265-4042-93f3-22a1117c4886",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "draw(pairs_graph)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "00bc2e82-9bc1-4119-b8cb-4a1c18afee19",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "***************************************************************************************************************************\n",
      "                                         Μethod:  Entity Matching\n",
      "***************************************************************************************************************************\n",
      "Method name: Entity Matching\n",
      "Parameters: \n",
      "\tTokenizer: white_space_tokenizer\n",
      "\tMetric: jaccard\n",
      "\tSimilarity Threshold: 0.5\n",
      "Runtime: 0.0726 seconds\n",
      "───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────\n",
      "Performance:\n",
      "\tPrecision:     97.67% \n",
      "\tRecall:         0.24%\n",
      "\tF1-score:       0.49%\n",
      "───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────\n"
     ]
    }
   ],
   "source": [
    "_ = EM.evaluate(pairs_graph)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "93b72120-4578-4d5c-a408-a24ee78bf6cb",
   "metadata": {},
   "source": [
    "## Entity Clustering\n",
    "\n",
    "It takes as input the similarity graph produced by Entity Matching and partitions it into a set of equivalence clusters, with every cluster corresponding to a distinct real-world object."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "500d2ef7-7017-4dba-bbea-acdba8abf5b7",
   "metadata": {},
   "outputs": [],
   "source": [
    "from pyjedai.clustering import ConnectedComponentsClustering"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "aebd9329-3a4b-48c9-bd05-c7bd4aed3ca9",
   "metadata": {},
   "outputs": [],
   "source": [
    "ec = ConnectedComponentsClustering()\n",
    "clusters = ec.process(pairs_graph, data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "3d2aa574",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "***************************************************************************************************************************\n",
      "                                         Μethod:  Connected Components Clustering\n",
      "***************************************************************************************************************************\n",
      "Method name: Connected Components Clustering\n",
      "Parameters: \n",
      "Runtime: 0.0000 seconds\n",
      "───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────\n",
      "Performance:\n",
      "\tPrecision:     97.92% \n",
      "\tRecall:         0.27%\n",
      "\tF1-score:       0.55%\n",
      "───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────\n"
     ]
    }
   ],
   "source": [
    "_ = ec.evaluate(clusters)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6e44642b-b0d9-4f8d-9fe4-4fca2ad716aa",
   "metadata": {},
   "source": [
    "# Workflow with Similarity Joins\n",
    "\n",
    "In this notebook we created the bellow archtecture:\n",
    "\n",
    "![workflow2-cora.png](https://github.com/AI-team-UoA/pyJedAI/blob/main/documentation/workflow2-cora.png?raw=true)\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2b07ea13-58cb-498a-949a-4702ea9ee4ce",
   "metadata": {
    "tags": []
   },
   "source": [
    "## Data Reading"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a0b7acf5-32b3-45b1-8347-197dfa869fb9",
   "metadata": {},
   "source": [
    "Data is the connecting module of all steps of the workflow"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "3006b051-8348-4922-a627-56441a1db7b7",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "from pyjedai.datamodel import Data\n",
    "d1 = pd.read_csv(\"./../data/der/cora/cora.csv\", sep='|')\n",
    "gt = pd.read_csv(\"./../data/der/cora/cora_gt.csv\", sep='|', header=None)\n",
    "attr = ['Entity Id','author', 'title']\n",
    "data = Data(\n",
    "    dataset_1=d1,\n",
    "    id_column_name_1='Entity Id',\n",
    "    ground_truth=gt,\n",
    "    attributes_1=attr\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b3eedb4c-b86f-4f98-abf2-9d6f7d0271c5",
   "metadata": {},
   "source": [
    "## Similarity Joins"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "afd97b7e-4bf8-4256-b9fc-a301e413e834",
   "metadata": {},
   "outputs": [],
   "source": [
    "from pyjedai.joins import ΕJoin, TopKJoin"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "7bc8ba43-b059-4839-8958-0b31fab95e46",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d86c40bfd439456a8cce134c648a89f2",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "EJoin (jaccard):   0%|          | 0/2590 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "join = ΕJoin(similarity_threshold = 0.5,\n",
    "             metric = 'jaccard',\n",
    "             tokenization = 'qgrams_multiset',\n",
    "             qgrams = 2)\n",
    "\n",
    "g = join.fit(data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "8fe34c67-4679-4279-8e7e-35437783ecf3",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "***************************************************************************************************************************\n",
      "                                         Μethod:  EJoin\n",
      "***************************************************************************************************************************\n",
      "Method name: EJoin\n",
      "Parameters: \n",
      "\tsimilarity_threshold: 0.5\n",
      "\tmetric: jaccard\n",
      "\ttokenization: qgrams_multiset\n",
      "\tqgrams: 2\n",
      "Runtime: 61.2715 seconds\n",
      "───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────\n",
      "Performance:\n",
      "\tPrecision:     65.80% \n",
      "\tRecall:        93.03%\n",
      "\tF1-score:      77.08%\n",
      "───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "{'Precision %': 65.79553872746729,\n",
       " 'Recall %': 93.03421787709497,\n",
       " 'F1 %': 77.07921508124006,\n",
       " 'True Positives': 15987,\n",
       " 'False Positives': 8311,\n",
       " 'True Negatives': 828357.0,\n",
       " 'False Negatives': 1197}"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "_ = join.evaluate(g)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "27c7bc36-c39b-4bab-a728-971195763728",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "62b551bae8764a019b65d4ff58bda29c",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Top-K Join (jaccard):   0%|          | 0/2590 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "topk_join = TopKJoin(K=20,\n",
    "             metric = 'jaccard',\n",
    "             tokenization = 'qgrams',\n",
    "             qgrams = 3)\n",
    "\n",
    "g = topk_join.fit(data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "e1061a7a-9ccb-41ec-8bdf-42412d4d249c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "draw(g)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "a4c57244-0c98-4598-8ae2-06cdd0c48385",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "***************************************************************************************************************************\n",
      "                                         Μethod:  Top-K Join\n",
      "***************************************************************************************************************************\n",
      "Method name: Top-K Join\n",
      "Parameters: \n",
      "\tsimilarity_threshold: 0.25547445255474455\n",
      "\tK: 20\n",
      "\tmetric: jaccard\n",
      "\ttokenization: qgrams\n",
      "\tqgrams: 3\n",
      "Runtime: 41.0031 seconds\n",
      "───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────\n",
      "Performance:\n",
      "\tPrecision:     58.34% \n",
      "\tRecall:        63.75%\n",
      "\tF1-score:      60.92%\n",
      "───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "{'Precision %': 58.340434597358325,\n",
       " 'Recall %': 63.74534450651769,\n",
       " 'F1 %': 60.923248053392655,\n",
       " 'True Positives': 10954,\n",
       " 'False Positives': 7822,\n",
       " 'True Negatives': 823813.0,\n",
       " 'False Negatives': 6230}"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "topk_join.evaluate(g)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e686ba55-cbbe-4133-8256-fa5339e70720",
   "metadata": {},
   "source": [
    "## Entity Clustering"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "e48627ef-0be6-45dd-b3ba-863818eadbfb",
   "metadata": {},
   "outputs": [],
   "source": [
    "from pyjedai.clustering import ConnectedComponentsClustering"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "a9d5a28d-79f0-479f-b154-f5f7e3661897",
   "metadata": {},
   "outputs": [],
   "source": [
    "ccc = ConnectedComponentsClustering()\n",
    "\n",
    "clusters = ccc.process(g, data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "f95539fe-2569-4569-8929-2f4220f14157",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "***************************************************************************************************************************\n",
      "                                         Μethod:  Connected Components Clustering\n",
      "***************************************************************************************************************************\n",
      "Method name: Connected Components Clustering\n",
      "Parameters: \n",
      "Runtime: 0.0630 seconds\n",
      "───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────\n",
      "Performance:\n",
      "\tPrecision:      2.05% \n",
      "\tRecall:       100.00%\n",
      "\tF1-score:       4.02%\n",
      "───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────\n"
     ]
    }
   ],
   "source": [
    "_ = ccc.evaluate(clusters)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "778b9b19-0964-4095-a293-a8336fe0607a",
   "metadata": {},
   "source": [
    "<hr>\n",
    "<div align=\"right\">\n",
    "K. Nikoletos, G. Papadakis & M. Koubarakis\n",
    "</div>\n",
    "<div align=\"right\">\n",
    "<a href=\"https://github.com/Nikoletos-K/pyJedAI/blob/main/LICENSE\">Apache License 2.0</a>\n",
    "</div>"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}