{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Node classification via node representations with attri2vec" ] }, { "cell_type": "markdown", "metadata": { "nbsphinx": "hidden", "tags": [ "CloudRunner" ] }, "source": [ "
Run the latest release of this notebook:
" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "This is the python implementation of how to perform node classification with the attri2vec algorithm outlined in paper [1]. The implementation uses the stellargraph components.\n", "\n", "\n", "**References:** \n", "\n", "[1] [Attributed Network Embedding via Subspace Discovery](https://link.springer.com/article/10.1007/s10618-019-00650-2). D. Zhang, Y. Jie, X. Zhu and C. Zhang, Data Mining and Knowledge Discovery, 2019. " ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## attri2vec\n", "\n", "attri2vec learns node representations by performing a linear/non-linear mapping on node content attributes. To make the learned node representations respect structural similarity, [DeepWalk](https://dl.acm.org/citation.cfm?id=2623732)/[Node2Vec](https://snap.stanford.edu/node2vec) learning mechanism is used to make nodes sharing similar random walk context nodes represented closely in the subspace. \n", "\n", "For each (``target``,``context``) node pair $(v_i,v_j)$ collected from random walks, attri2vec learns the representation for the target node $v_i$ by using it to predict the existence of context node $v_j$, with the following three-layer neural network." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "![](attri2vec-illustration.png)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Node $v_i$'s representation in the hidden layer is obtained by multiplying $v_i$'s raw content feature vector in the input layer with the input-to-hidden weight matrix $W_{in}$ followed by an activation function. The existence probability of each node conditioned on node $v_i$ is outputted in the output layer, which is obtained by multiplying $v_i$'s hidden-layer representation with the hidden-to-out weight matrix $W_{out}$ followed by a softmax activation. To capture the ``target-context`` relation between $v_i$ and $v_j$, we need to maximize the probability $\\mathrm{P}(v_j|v_i)$. However, computing $\\mathrm{P}(v_j|v_i)$ is time consuming, which involves the matrix multiplication between $v_i$'s hidden-layer representation and the hidden-to-out weight matrix $W_{out}$." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "To speed up the computing, we adopt the negative sampling strategy [1]. For each (``target``, ``context``) node pair, we sample a negative node $v_k$, which is not $v_i$'s context. To obtain the output, instead of multiplying $v_i$'s hidden-layer representation with the hidden-to-out weight matrix $W_{out}$ followed by a softmax activation, we only calculate the dot product between $v_i$'s hidden-layer representation and the $j$th column as well as the $k$th column of the hidden-to-output weight matrix $W_{out}$ followed by a sigmoid activation respectively. According to [1], the original objective to maximize $\\mathrm{P}(v_j|v_i)$ can be approximated by minimizing the cross entropy between $v_j$ and $v_k$'s outputs and their ground-truth labels (1 for $v_j$ and 0 for $v_k$)." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "The entire model is trained end-to-end by minimizing the binary cross-entropy loss function with regards to predicted node pair labels and true node pair labels, using stochastic gradient descent (SGD) updates of the model parameters, with minibatches of 'training' node pairs generated on demand and fed into the model." ] }, { "cell_type": "code", "execution_count": 1, "metadata": { "nbsphinx": "hidden", "tags": [ "CloudRunner" ] }, "outputs": [], "source": [ "# install StellarGraph if running on Google Colab\n", "import sys\n", "if 'google.colab' in sys.modules:\n", " %pip install -q stellargraph[demos]==1.2.1" ] }, { "cell_type": "code", "execution_count": 2, "metadata": { "nbsphinx": "hidden", "tags": [ "VersionCheck" ] }, "outputs": [], "source": [ "# verify that we're using the correct version of StellarGraph for this notebook\n", "import stellargraph as sg\n", "\n", "try:\n", " sg.utils.validate_notebook_version(\"1.2.1\")\n", "except AttributeError:\n", " raise ValueError(\n", " f\"This notebook requires StellarGraph version 1.2.1, but a different version {sg.__version__} is installed. Please see .\"\n", " ) from None" ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [], "source": [ "import networkx as nx\n", "import pandas as pd\n", "import numpy as np\n", "import os\n", "import random\n", "\n", "import stellargraph as sg\n", "from stellargraph.data import UnsupervisedSampler\n", "from stellargraph.mapper import Attri2VecLinkGenerator, Attri2VecNodeGenerator\n", "from stellargraph.layer import Attri2Vec, link_classification\n", "\n", "from tensorflow import keras\n", "\n", "from pandas.core.indexes.base import Index\n", "\n", "import matplotlib.pyplot as plt\n", "from sklearn.manifold import TSNE\n", "from sklearn.decomposition import PCA\n", "\n", "from sklearn.model_selection import train_test_split\n", "from sklearn.linear_model import LogisticRegressionCV\n", "from sklearn.metrics import accuracy_score\n", "from stellargraph import datasets\n", "from IPython.display import display, HTML" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Dataset\n", "\n", "For clarity we ignore isolated nodes and subgraphs and use only the largest connected component." ] }, { "cell_type": "markdown", "metadata": { "tags": [ "DataLoadingLinks" ] }, "source": [ "(See [the \"Loading from Pandas\" demo](../basics/loading-pandas.ipynb) for details on how data can be loaded.)" ] }, { "cell_type": "code", "execution_count": 4, "metadata": { "tags": [ "DataLoading" ] }, "outputs": [ { "data": { "text/html": [ "The CiteSeer dataset consists of 3312 scientific publications classified into one of six classes. The citation network consists of 4732 links, although 17 of these have a source or target publication that isn't in the dataset and only 4715 are included in the graph. Each publication in the dataset is described by a 0/1-valued word vector indicating the absence/presence of the corresponding word from the dictionary. The dictionary consists of 3703 unique words." ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "dataset = datasets.CiteSeer()\n", "display(HTML(dataset.description))\n", "G, subjects = dataset.load(largest_connected_component_only=True)" ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "StellarGraph: Undirected multigraph\n", " Nodes: 2110, Edges: 3757\n", "\n", " Node types:\n", " paper: [2110]\n", " Features: float32 vector, length 3703\n", " Edge types: paper-cites->paper\n", "\n", " Edge types:\n", " paper-cites->paper: [3757]\n", " Weights: all 1 (default)\n", " Features: none\n" ] } ], "source": [ "print(G.info())" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Train attri2vec on Citeseer" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Specify the other optional parameter values: root nodes, the number of walks to take per node, the length of each walk." ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [], "source": [ "nodes = list(G.nodes())\n", "number_of_walks = 4\n", "length = 5" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Create the UnsupervisedSampler instance with the relevant parameters passed to it." ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [], "source": [ "unsupervised_samples = UnsupervisedSampler(\n", " G, nodes=nodes, length=length, number_of_walks=number_of_walks\n", ")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Set the batch size and the number of epochs. " ] }, { "cell_type": "code", "execution_count": 8, "metadata": {}, "outputs": [], "source": [ "batch_size = 50\n", "epochs = 4" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Define an attri2vec generator, which generates batches of (target, context) nodes and labels for the node pair." ] }, { "cell_type": "code", "execution_count": 9, "metadata": {}, "outputs": [], "source": [ "generator = Attri2VecLinkGenerator(G, batch_size)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Building the model: a 1-hidden-layer node representation ('input embedding') of the `target` node and the parameter vector ('output embedding') for predicting the existence of `context node` for each `(target context)` pair, with a link classification layer performed on the dot product of the 'input embedding' of the `target` node and the 'output embedding' of the `context` node.\n", "\n", "Attri2Vec part of the model, with a 128-dimension hidden layer, no bias term and no normalization. (Normalization can be set to 'l2'). " ] }, { "cell_type": "code", "execution_count": 10, "metadata": {}, "outputs": [], "source": [ "layer_sizes = [128]\n", "attri2vec = Attri2Vec(\n", " layer_sizes=layer_sizes, generator=generator, bias=False, normalize=None\n", ")" ] }, { "cell_type": "code", "execution_count": 11, "metadata": {}, "outputs": [], "source": [ "# Build the model and expose input and output sockets of attri2vec, for node pair inputs:\n", "x_inp, x_out = attri2vec.in_out_tensors()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Use the link_classification function to generate the prediction, with the `ip` edge embedding generation method and the `sigmoid` activation, which actually performs the dot product of the 'input embedding' of the target node and the 'output embedding' of the context node followed by a sigmoid activation. " ] }, { "cell_type": "code", "execution_count": 12, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "link_classification: using 'ip' method to combine node embeddings into edge embeddings\n" ] } ], "source": [ "prediction = link_classification(\n", " output_dim=1, output_act=\"sigmoid\", edge_embedding_method=\"ip\"\n", ")(x_out)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Stack the Attri2Vec encoder and prediction layer into a Keras model, and specify the loss." ] }, { "cell_type": "code", "execution_count": 13, "metadata": {}, "outputs": [], "source": [ "model = keras.Model(inputs=x_inp, outputs=prediction)\n", "\n", "model.compile(\n", " optimizer=keras.optimizers.Adam(lr=1e-3),\n", " loss=keras.losses.binary_crossentropy,\n", " metrics=[keras.metrics.binary_accuracy],\n", ")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Train the model." ] }, { "cell_type": "code", "execution_count": 14, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Train for 1351 steps\n", "Epoch 1/4\n", "1351/1351 - 6s - loss: 0.6822 - binary_accuracy: 0.5551\n", "Epoch 2/4\n", "1351/1351 - 5s - loss: 0.5173 - binary_accuracy: 0.7547\n", "Epoch 3/4\n", "1351/1351 - 5s - loss: 0.3163 - binary_accuracy: 0.8961\n", "Epoch 4/4\n", "1351/1351 - 5s - loss: 0.2059 - binary_accuracy: 0.9439\n" ] } ], "source": [ "history = model.fit(\n", " generator.flow(unsupervised_samples),\n", " epochs=epochs,\n", " verbose=2,\n", " use_multiprocessing=False,\n", " workers=1,\n", " shuffle=True,\n", ")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Visualise Node Embeddings" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Build the node based model for predicting node representations from node content attributes with the learned parameters. Below a Keras model is constructed, with `x_inp[0]` as input and `x_out[0]` as output. Note that this model's weights are the same as those of the corresponding node encoder in the previously trained node pair classifier." ] }, { "cell_type": "code", "execution_count": 15, "metadata": {}, "outputs": [], "source": [ "x_inp_src = x_inp[0]\n", "x_out_src = x_out[0]\n", "embedding_model = keras.Model(inputs=x_inp_src, outputs=x_out_src)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Get the node embeddings by applying the learned mapping function to node content features." ] }, { "cell_type": "code", "execution_count": 16, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "43/43 [==============================] - 0s 2ms/step\n" ] } ], "source": [ "node_gen = Attri2VecNodeGenerator(G, batch_size).flow(subjects.index)\n", "node_embeddings = embedding_model.predict(node_gen, workers=1, verbose=1)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Transform the embeddings to 2d space for visualisation." ] }, { "cell_type": "code", "execution_count": 17, "metadata": {}, "outputs": [], "source": [ "transform = TSNE # PCA\n", "\n", "trans = transform(n_components=2)\n", "node_embeddings_2d = trans.fit_transform(node_embeddings)" ] }, { "cell_type": "code", "execution_count": 18, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "# draw the embedding points, coloring them by the target label (paper subject)\n", "alpha = 0.7\n", "label_map = {l: i for i, l in enumerate(np.unique(subjects))}\n", "node_colours = [label_map[target] for target in subjects]\n", "\n", "plt.figure(figsize=(7, 7))\n", "plt.axes().set(aspect=\"equal\")\n", "plt.scatter(\n", " node_embeddings_2d[:, 0],\n", " node_embeddings_2d[:, 1],\n", " c=node_colours,\n", " cmap=\"jet\",\n", " alpha=alpha,\n", ")\n", "plt.title(\"{} visualization of node embeddings\".format(transform.__name__))\n", "plt.show()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Node Classification Task" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "The embeddings learned by `attri2vec` can be used as feature vectors in downstream tasks, such as node classification and link prediction.\n", "\n", "In this example, we will use the `attri2vec` node embeddings to train a classifier to predict the subject of a paper in DBLP." ] }, { "cell_type": "code", "execution_count": 19, "metadata": {}, "outputs": [], "source": [ "# X will hold the 128-dimensional input features\n", "X = node_embeddings\n", "# y holds the corresponding target values\n", "y = np.array(subjects)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Data Splitting\n", "\n", "We split the data into train and test sets. \n", "\n", "We use 20% of the data for training and the remaining 80% for testing as a hold out test set." ] }, { "cell_type": "code", "execution_count": 20, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Array shapes:\n", " X_train = (422, 128)\n", " y_train = (422,)\n", " X_test = (1688, 128)\n", " y_test = (1688,)\n" ] } ], "source": [ "X_train, X_test, y_train, y_test = train_test_split(X, y, train_size=0.2, test_size=None)\n", "print(\n", " \"Array shapes:\\n X_train = {}\\n y_train = {}\\n X_test = {}\\n y_test = {}\".format(\n", " X_train.shape, y_train.shape, X_test.shape, y_test.shape\n", " )\n", ")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Classifier Training\n", "\n", "We train a Logistic Regression classifier on the training data. " ] }, { "cell_type": "code", "execution_count": 21, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "LogisticRegressionCV(Cs=10, class_weight=None, cv=10, dual=False,\n", " fit_intercept=True, intercept_scaling=1.0, l1_ratios=None,\n", " max_iter=1000, multi_class='ovr', n_jobs=None,\n", " penalty='l2', random_state=None, refit=True,\n", " scoring='accuracy', solver='lbfgs', tol=0.0001,\n", " verbose=False)" ] }, "execution_count": 21, "metadata": {}, "output_type": "execute_result" } ], "source": [ "clf = LogisticRegressionCV(\n", " Cs=10, cv=10, scoring=\"accuracy\", verbose=False, multi_class=\"ovr\", max_iter=1000\n", ")\n", "clf.fit(X_train, y_train)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Predict the hold-out test set." ] }, { "cell_type": "code", "execution_count": 22, "metadata": {}, "outputs": [], "source": [ "y_pred = clf.predict(X_test)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Calculate the accuracy of the classifier on the test set." ] }, { "cell_type": "code", "execution_count": 23, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "0.7535545023696683" ] }, "execution_count": 23, "metadata": {}, "output_type": "execute_result" } ], "source": [ "accuracy_score(y_test, y_pred)" ] }, { "cell_type": "markdown", "metadata": { "nbsphinx": "hidden", "tags": [ "CloudRunner" ] }, "source": [ "
Run the latest release of this notebook:
" ] } ], "metadata": { "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": "3.6.9" } }, "nbformat": 4, "nbformat_minor": 4 }