A survey of structural representation learning for social networks

Social networks have a plethora of applications, and analysis of these applications has been gaining much interest from the research community. The high dimensionality of social network data poses a significant obstacle in its analysis, leading to the curse of dimensionality. The mushrooming of representation learning in various research fields facilitates network representation learning (also called network embedding), which will help us address the above-mentioned issue. Structural Representation Learning aims to learn low-dimensional vector representations of high-dimensional network data, allowing maximal preservation of network structural information. This representation can then serve as a backbone for various network-based applications. First, we investigate the techniques used in network representation learning and similarity indices. We then categorize the representative algorithms into three types based on the network structural level used in their learning process. We also introduce algorithms for representation learning of edges, subgraphs, and the whole network. Finally, we introduce the evaluation metrics and the applications of network representation learning and promising future research directions. (C) 2022 Elsevier B.V. All rights reserved.

Automated summary

Social networks have a wide range of applications, and the analysis of these applications has attracted a lot of attention from the research community. However, the high dimensionality of social network data presents a challenge in its analysis, known as the curse of dimensionality. Representation learning offers a solution by learning low-dimensional vector representations of high-dimensional network data while preserving network structural information. These representations can be utilized in various network-based applications.