Resampled Mutual Information for Clustering and Community Detection

2024/11/21 by Lim, Cheaheon
Computer Science · Physics and Astronomy · #Text and Document Classification Technologies #Web Data Mining and Analysis #Complex Network Analysis Techniques

paper · pdf · doi:10.48550/arxiv.2412.03584

Abstract

We introduce resampled mutual information (ResMI), a novel measure of clustering similarity that combines insights from information theoretic and pair counting approaches to clustering and community detection. Similar to chance-corrected measures, ResMI satisfies the constant baseline property, but it has the advantages of not requiring adjustment terms and being fully interpretable in the language of information theory. Experiments on synthetic datasets demonstrate that ResMI is robust to common biases exhibited by existing measures, particularly in settings with high cluster counts and asymmetric cluster distributions. Additionally, we show that ResMI identifies meaningful community structures in two real contact tracing networks.

Related

• Normalized Mutual Information to evaluate overlapping community finding algorithms
• Normalized mutual information is a biased measure for classification and community detection
• Mutual Community Detection across Multiple Partially Aligned Social Networks
• Evaluation of Community Detection Methods
• Pairwise Adjusted Mutual Information