Model-based clustering for random hypergraphs

TL;DR

This paper introduces a probabilistic model for clustering in random hypergraphs, extending latent class analysis to handle higher-order interactions, with an EM algorithm for parameter estimation and model selection, demonstrated on real data.

Contribution

It presents a novel probabilistic hypergraph clustering model extending latent class analysis, with an EM algorithm and model selection method, applied to real-world data.

Findings

Effective clustering of real-world hypergraph data

Model outperforms existing methods in capturing higher-order interactions

Demonstrates practical applicability on three datasets

Abstract

A probabilistic model for random hypergraphs is introduced to represent unary, binary and higher order interactions among objects in real-world problems. This model is an extension of the Latent Class Analysis model, which captures clustering structures among objects. An EM (expectation maximization) algorithm with MM (minorization maximization) steps is developed to perform parameter estimation while a cross validated likelihood approach is employed to perform model selection. The developed model is applied to three real-world data sets where interesting results are obtained.