Synwalk -- Community Detection via Random Walk Modelling

TL;DR

Synwalk is a robust, random walk-based community detection method that synthesizes multiple candidate walks to effectively identify communities across diverse network types, outperforming some existing algorithms in certain scenarios.

Contribution

Introduces Synwalk, a novel community detection approach that synthesizes random walks, with solid theoretical foundations and validated performance on synthetic and real networks.

Findings

Synwalk performs robustly across networks with varying parameters.

Outperforms Infomap on high mixing parameter networks.

Matches or exceeds Walktrap and Infomap on networks with small communities.

Abstract

Complex systems, abstractly represented as networks, are ubiquitous in everyday life. Analyzing and understanding these systems requires, among others, tools for community detection. As no single best community detection algorithm can exist, robustness across a wide variety of problem settings is desirable. In this work, we present Synwalk, a random walk-based community detection method. Synwalk builds upon a solid theoretical basis and detects communities by synthesizing the random walk induced by the given network from a class of candidate random walks. We thoroughly validate the effectiveness of our approach on synthetic and empirical networks, respectively, and compare Synwalk's performance with the performance of Infomap and Walktrap. Our results indicate that Synwalk performs robustly on networks with varying mixing parameters and degree distributions. We outperform Infomap on networks with high mixing parameter, and Infomap and Walktrap on networks with many small communities and low average degree. Our work has a potential to inspire further development of community detection via synthesis of random walks and we provide concrete ideas for future research.