Modularity-Aware Graph Autoencoders for Joint Community Detection and Link Prediction

TL;DR

This paper introduces Modularity-Aware GAE and VGAE models that jointly improve community detection and link prediction by incorporating modularity-based priors and novel training strategies, outperforming existing methods.

Contribution

It proposes a community-preserving message passing scheme and a modularity-inspired regularizer to enhance GAE and VGAE for joint community detection and link prediction.

Findings

Effective in improving community detection accuracy

Maintains strong link prediction performance

Validated on various real-world graphs

Abstract

Graph autoencoders (GAE) and variational graph autoencoders (VGAE) emerged as powerful methods for link prediction. Their performances are less impressive on community detection problems where, according to recent and concurring experimental evaluations, they are often outperformed by simpler alternatives such as the Louvain method. It is currently still unclear to which extent one can improve community detection with GAE and VGAE, especially in the absence of node features. It is moreover uncertain whether one could do so while simultaneously preserving good performances on link prediction. In this paper, we show that jointly addressing these two tasks with high accuracy is possible. For this purpose, we introduce and theoretically study a community-preserving message passing scheme, doping our GAE and VGAE encoders by considering both the initial graph structure and modularity-based prior communities when computing embedding spaces. We also propose novel training and optimization strategies, including the introduction of a modularity-inspired regularizer complementing the existing reconstruction losses for joint link prediction and community detection. We demonstrate the empirical effectiveness of our approach, referred to as Modularity-Aware GAE and VGAE, through in-depth experimental validation on various real-world graphs.