Efficient inference of overlapping communities in complex networks

TL;DR

This paper introduces a Bayesian multiple-networks stochastic blockmodel for efficient overlapping community detection in complex networks, improving structure inference and link prediction accuracy.

Contribution

It formulates a novel generative Bayesian framework for separating networks into subnetworks, enabling easier structure inference and avoiding complex group interaction definitions.

Findings

Successful recovery of planted structures in synthetic networks

Enhanced link prediction performance on real-world datasets

Efficient parameter inference using Gibbs sampling

Abstract

We discuss two views on extending existing methods for complex network modeling which we dub the communities first and the networks first view, respectively. Inspired by the networks first view that we attribute to White, Boorman, and Breiger (1976)[1], we formulate the multiple-networks stochastic blockmodel (MNSBM), which seeks to separate the observed network into subnetworks of different types and where the problem of inferring structure in each subnetwork becomes easier. We show how this model is specified in a generative Bayesian framework where parameters can be inferred efficiently using Gibbs sampling. The result is an effective multiple-membership model without the drawbacks of introducing complex definitions of "groups" and how they interact. We demonstrate results on the recovery of planted structure in synthetic networks and show very encouraging results on link prediction performances using multiple-networks models on a number of real-world network data sets.