Community Detection from Multiple Observations: from Product Graph Model to Brain Applications

TL;DR

This paper introduces a multilayer graph model for community detection from multiple observations, demonstrating improved accuracy over existing methods and applying it to EEG brain network analysis during motor imagery tasks.

Contribution

The paper develops a novel multilayer graph approach based on the Cartesian product and Laplacian eigenstructure for community detection from multiple graph observations.

Findings

Outperforms state-of-the-art in synthetic graph community detection

Accurately identifies brain network communities from EEG data

Shows promise for EEG-based motor imagery applications

Abstract

This paper proposes a multilayer graph model for the community detection from multiple observations. This is a very frequent situation, when different estimators are applied to infer graph edges from signals at its nodes, or when different signal measurements are carried out. The multilayer network stacks the graph observations at the different layers, and it links replica nodes at adjacent layers. This configuration matches the Cartesian product between the ground truth graph and a path graph, where the number of nodes corresponds to the number of the observations. Stemming on the algebraic structure of the Laplacian of the Cartesian multilayer network, we infer a subset of the eigenvectors of the true graph and perform community detection. Experimental results on synthetic graphs prove the accuracy of the method, which outperforms state-of-the-art approaches in terms of ability of correctly detecting graph communities. Finally, we show the application of our method to discriminate different brain networks derived from real EEG data collected during motor imagery experiments. We conclude that our approach appears promising in identifying graph communities when multiple observations of the graph are available and it results promising for EEG-based motor imagery applications.