Machine Learning in Social Networks: Embedding Nodes, Edges, Communities, and Graphs

Introduction

1.1 introduction

1.2 Notations used in Book

1.3 Contents covered in this book

2 Representations of Networks

2.1 Introduction

2.2 Networks Represented as Graphs

2.3 Data Structures to Represent Graphs

2.3.1 Matrix Representation

2.4 Network Embeddings

2.5 Evaluation Datasets

2.5.1 Evaluation Datasets

2.5.2 Evaluation Metrics

2.6 Machine Learning Downstream Tasks

2.6.1 Classification

2.6.3 Link Prediction (LP)

2.6.4 Visualization

2.6.5 Network Reconstruction

2.7 Embeddings based on Matrix Factorization

2.7.1 Singular Value Decomposition (SVD)

2.7.2 Matrix Factorization based Clustering

2.7.4 Non-negative Matrix factorization (NMF)

2.8 Word2vec

2.8.1 Skipgram model

2.9 Learning Network Embeddings

2.9.1 Supervised Learning

2.9.2 Unsupervised Learning

2.9.4 Graph Embedding

2.10 Summary

3 Deep Learning

3.1 Introduction

3.2 Neural Networks

3.2.1 Perceptron

3.2.2 Characteristics of Neural Networks

3.2.3 Multilayer Perceptron Networks

3.2.4 Training MLP Networks

3.3.1 Activation Function

3.3.2 Initialization of Weights

3.3.3 Deep Feedforward Neural Network

3.4 Recurrent Networks

3.4.1 Recurrent Neural Networks

3.4.2 Long Short Term Memory

3.4.4 Training of LSTM Models

3.5 Learning Representations using Autoencoders

3.5.1 Types of Autoencoders

3.6 Summary

References

4 Embedding Nodes and Edge

4.1 Introduction

4.2 Representation of Node and Edges as Vectors

4.3 Embeddings based on Random Walks

4.5 Graph Neural Network Models

4.6 State of the art algorithms

4.7 Evaluation methods and Machine Learning tasks

4.8 Summary

References

5 Embedding Graphs

5.1 Introduction

5.2 Representation of Graphs as Vectors

5.3 Graph Representation using Node Embeddings

5.4.1 Global Pooling Methods

5.4.2 Hierarchical Pooling Methods

5.5 State of the art algorithms

5.6 Evaluation methods and Machine Learning tasks

5.7 Summary