Latent Space Approaches to Community Detection in Dynamic Networks

TL;DR

This paper introduces two novel Bayesian methods for community detection in dynamic networks using latent space models, effectively capturing temporal changes, directionality, transitivity, and individual propensities.

Contribution

It extends latent space community detection to dynamic networks with Bayesian algorithms, incorporating temporal, directed, and transitivity features.

Findings

Successfully applied to friendship and trade networks

Captured temporal community evolution

Demonstrated effectiveness of Bayesian estimation

Abstract

Embedding dyadic data into a latent space has long been a popular approach to modeling networks of all kinds. While clustering has been done using this approach for static networks, this paper gives two methods of community detection within dynamic network data, building upon the distance and projection models previously proposed in the literature. Our proposed approaches capture the time-varying aspect of the data, can model directed or undirected edges, inherently incorporate transitivity and account for each actor's individual propensity to form edges. We provide Bayesian estimation algorithms, and apply these methods to a ranked dynamic friendship network and world export/import data.