Ensemble-Based Algorithms to Detect Disjoint and Overlapping Communities in Networks

TL;DR

This paper introduces two ensemble algorithms, EnDisCO and MeDOC, for detecting disjoint and overlapping communities in networks, demonstrating superior performance over existing methods through extensive experiments.

Contribution

It presents novel ensemble algorithms for community detection, including the first ensemble method for overlapping communities, outperforming state-of-the-art approaches.

Findings

Both algorithms outperform non-ensemble methods significantly.

EnDisCO effectively identifies disjoint communities.

MeDOC successfully detects overlapping communities.

Abstract

Given a set ${\cal AL}$ of community detection algorithms and a graph $G$ as inputs, we propose two ensemble methods $\mathtt{EnDisCO}$ and $\mathtt{MeDOC}$ that (respectively) identify disjoint and overlapping communities in $G$. $\mathtt{EnDisCO}$ transforms a graph into a latent feature space by leveraging multiple base solutions and discovers disjoint community structure. $\mathtt{MeDOC}$ groups similar base communities into a meta-community and detects both disjoint and overlapping community structures. Experiments are conducted at different scales on both synthetically generated networks as well as on several real-world networks for which the underlying ground-truth community structure is available. Our extensive experiments show that both algorithms outperform state-of-the-art non-ensemble algorithms by a significant margin. Moreover, we compare $\mathtt{EnDisCO}$ and $\mathtt{MeDOC}$ with a recent ensemble method for disjoint community detection and show that our approaches achieve superior performance. To the best of our knowledge, $\mathtt{MeDOC}$ is the first ensemble approach for overlapping community detection.