Advances in Scaling Community Discovery Methods for Signed Graph Networks

TL;DR

This paper reviews and compares current community detection methods for signed networks, focusing on their effectiveness and scalability across various real-world datasets, and offers recommendations for future improvements.

Contribution

It provides a comprehensive evaluation of state-of-the-art algorithms for signed graph community detection, highlighting their strengths, limitations, and scalability issues.

Findings

Spectral clustering effectiveness varies with signed graph structure

Scalability challenges identified for large, dense signed networks

Recommendations for extending current methods are proposed

Abstract

Community detection is a common task in social network analysis (SNA) with applications in a variety of fields including medicine, criminology, and business. Despite the popularity of community detection, there is no clear consensus on the most effective methodology for signed networks. In this paper, we summarize the development of community detection in signed networks and evaluate current state-of-the-art techniques on several real-world data sets. First, we give a comprehensive background of community detection in signed graphs. Next, we compare various adaptations of the Laplacian matrix in recovering ground-truth community labels via spectral clustering in small signed graph data sets. Then, we evaluate the scalability of leading algorithms on small, large, dense, and sparse real-world signed graph networks. We conclude with a discussion of our novel findings and recommendations for extensions and improvements in state-of-the-art techniques for signed graph community discovery in real-world signed graphs.} {Sign Graph Clustering, Community Discovery, Sparse Networks