Latent Modularity in Multi-View Data

2025-11-01

AI Summary

This paper addresses multi-view clustering by proposing a “latent modularity” structure: it assumes a shared baseline clustering across individuals, while allowing view-specific deviations conditioned on this baseline to flexibly capture cross-view heterogeneity. Methodologically, we develop a Bayesian hierarchical model incorporating a latent modularity prior, derive the marginal prior distribution of view-specific cluster assignments, and thereby reveal the model’s intrinsic statistical properties; joint inference is performed via MCMC, enabling view-specific yet coupled clustering mechanisms. Experiments demonstrate that the model accurately recovers complex multi-view clustering structures, significantly outperforming state-of-the-art methods on both synthetic and real-world datasets—particularly in characterizing dynamic, view-dependent membership patterns of individuals.

Technology Category

• Machine Learning: Multi-instance/Multi-view Learning
• Computer Vision: Multi-modal Vision
• Reasoning under Uncertainty: Relational Probabilistic Models

Application Category

• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

In this article, we consider the problem of clustering multi-view data, that is, information associated to individuals that form heterogeneous data sources (the views). We adopt a Bayesian model and in the prior structure we assume that each individual belongs to a baseline cluster and conditionally allow each individual in each view to potentially belong to different clusters than the baseline. We call such a structure ''latent modularity''. Then for each cluster, in each view we have a specific statistical model with an associated prior. We derive expressions for the marginal priors on the view-specific cluster labels and the associated partitions, giving several insights into our chosen prior structure. Using simple Markov chain Monte Carlo algorithms, we consider our model in a simulation study, along with a more detailed case study that requires several modeling innovations.

Problem

• Clustering multi-view data from heterogeneous sources
• Modeling individual cluster memberships across different views
• Developing Bayesian prior structures for view-specific partitions

Innovation

• Bayesian model with baseline cluster assignment
• Latent modularity for view-specific clustering
• Markov chain Monte Carlo for model inference