Component models for large networks

TL;DR

This paper introduces the interaction component model for communities (ICMc), a scalable Bayesian approach for large networks that captures community structures and uncertainty, demonstrated on a massive social network dataset.

Contribution

It presents a novel interaction component model for networks using Dirichlet Process priors, enabling scalable Bayesian community detection with uncertainty quantification.

Findings

Successfully applied to a network with 670,000 nodes and 1.89 million links

Finds both community-like and disassortative structures

Demonstrates high scalability and efficiency

Abstract

Being among the easiest ways to find meaningful structure from discrete data, Latent Dirichlet Allocation (LDA) and related component models have been applied widely. They are simple, computationally fast and scalable, interpretable, and admit nonparametric priors. In the currently popular field of network modeling, relatively little work has taken uncertainty of data seriously in the Bayesian sense, and component models have been introduced to the field only recently, by treating each node as a bag of out-going links. We introduce an alternative, interaction component model for communities (ICMc), where the whole network is a bag of links, stemming from different components. The former finds both disassortative and assortative structure, while the alternative assumes assortativity and finds community-like structures like the earlier methods motivated by physics. With Dirichlet Process priors and an efficient implementation the models are highly scalable, as demonstrated with a social network from the Last.fm web site, with 670,000 nodes and 1.89 million links.