Machine learning in social networks. Embedding nodes, edges, communities, and graphs (Q2218679): Difference between revisions

This book introduces network representation learning in social networks in a concise manner. It gives an interesting elementary description for methods and concepts used in learning representation and social network analysis, which is suitable for undergraduate level readers. The book begins with an introduction to social network embedding by using an example in network representation learning. Chapter 2 describes different methods for representation of networks and focuses on the machine learning downstream task and metrics. The embedding approaches based on matrix factorization are discussed. Chapter 3 covers the ideas and frameworks of neural networks, convolutional neural networks and recurrent networks. Different types of auto encoders are discussed. Chapter 4 studies node representation methods including random walks, matrix factorization, and graph neural networks. Chapter 5 studies various strategies for graph embeddings including SortPool, DIFFPOOL, SAGPool and GIN etc. The book culminates in Chapter 6 with some useful conclusions. Each chapter features a bibliography and a summary for the recap of main points. The book is a quick introduction to machine learning topics in social network analysis.