Network representation learning: A macro and micro view

TL;DR

This survey comprehensively reviews the development of network representation learning, categorizing algorithms into shallow, heterogeneous, and graph neural network models, and analyzing their theoretical foundations.

Contribution

It provides a systematic classification and comparison of current algorithms, offering deep insights into their underlying theories and development trends.

Findings

Categorizes network embedding algorithms into three main groups.

Analyzes the theoretical foundations of different embedding models.

Highlights the evolution and future directions of the field.

Abstract

Graph is a universe data structure that is widely used to organize data in real-world. Various real-word networks like the transportation network, social and academic network can be represented by graphs. Recent years have witnessed the quick development on representing vertices in the network into a low-dimensional vector space, referred to as network representation learning. Representation learning can facilitate the design of new algorithms on the graph data. In this survey, we conduct a comprehensive review of current literature on network representation learning. Existing algorithms can be categorized into three groups: shallow embedding models, heterogeneous network embedding models, graph neural network based models. We review state-of-the-art algorithms for each category and discuss the essential differences between these algorithms. One advantage of the survey is that we systematically study the underlying theoretical foundations underlying the different categories of algorithms, which offers deep insights for better understanding the development of the network representation learning field.