Unsupervised Community Detection with Modularity-Based Attention Model

TL;DR

This paper introduces an unsupervised community detection method using a modularity-based attention model that effectively clusters nodes in graphs, competing with classical and GNN approaches, and is trainable on a single graph.

Contribution

It presents a novel unsupervised algorithm that encodes Bethe Hessian embeddings via soft modularity loss, applying attention models to graph clustering in a hard regime.

Findings

Competitive performance with classical and GNN models

Can be trained on a single graph

Effective in hard clustering regimes

Abstract

In this paper we take a problem of unsupervised nodes clustering on graphs and show how recent advances in attention models can be applied successfully in a "hard" regime of the problem. We propose an unsupervised algorithm that encodes Bethe Hessian embeddings by optimizing soft modularity loss and argue that our model is competitive to both classical and Graph Neural Network (GNN) models while it can be trained on a single graph.