Strategies for online inference of model-based clustering in large and growing networks

TL;DR

This paper introduces online inference strategies for model-based clustering in large networks, utilizing SAEM and variational methods, and demonstrates their effectiveness on simulated and real political network data.

Contribution

It adapts online estimation algorithms for large-scale network clustering, providing a faster and more precise alternative to existing methods.

Findings

Online EM algorithms achieve a good balance of speed and accuracy.

Methods successfully decipher political network structures.

Algorithms outperform some existing approaches in large network scenarios.

Abstract

In this paper we adapt online estimation strategies to perform model-based clustering on large networks. Our work focuses on two algorithms, the first based on the SAEM algorithm, and the second on variational methods. These two strategies are compared with existing approaches on simulated and real data. We use the method to decipher the connexion structure of the political websphere during the US political campaign in 2008. We show that our online EM-based algorithms offer a good trade-off between precision and speed, when estimating parameters for mixture distributions in the context of random graphs.