Mixtures of matrix Gaussian distributions provide a probabilistic framework for clustering continuous matrix-variate data, which are increasingly common in various fields. Despite their widespread use and successful applications, these models suffer from over-parameterization, making them not suitable for even moderately sized matrix-variate data. To address this issue, we introduce a sparse model-based clustering approach for three-way data. Our approach assumes that the matrix mixture parameters are sparse and have different degrees of sparsity across clusters, enabling the induction of parsimony in a flexible manner. Estimation relies on the maximization of a penalized likelihood, with specifically tailored group and graphical lasso penalties. These penalties facilitate the selection of the most informative features for clustering three-way data where variables are recorded over multiple occasions, as well as allowing the identification of cluster-specific association structures. We conduct extensive testing of the proposed methodology on synthetic data and validate its effectiveness through an application to time-dependent crime patterns across multiple U.S. cities. Supplementary files for this article are available online.

Sparse model-based clustering of three-way data via lasso-type penalties / A. Cappozzo, A. Casa, M. Fop. - In: JOURNAL OF COMPUTATIONAL AND GRAPHICAL STATISTICS. - ISSN 1061-8600. - (2024), pp. 1-71. [Epub ahead of print] [10.1080/10618600.2024.2429705]

Sparse model-based clustering of three-way data via lasso-type penalties

A. Cappozzo
Primo
;
2024

Abstract

Mixtures of matrix Gaussian distributions provide a probabilistic framework for clustering continuous matrix-variate data, which are increasingly common in various fields. Despite their widespread use and successful applications, these models suffer from over-parameterization, making them not suitable for even moderately sized matrix-variate data. To address this issue, we introduce a sparse model-based clustering approach for three-way data. Our approach assumes that the matrix mixture parameters are sparse and have different degrees of sparsity across clusters, enabling the induction of parsimony in a flexible manner. Estimation relies on the maximization of a penalized likelihood, with specifically tailored group and graphical lasso penalties. These penalties facilitate the selection of the most informative features for clustering three-way data where variables are recorded over multiple occasions, as well as allowing the identification of cluster-specific association structures. We conduct extensive testing of the proposed methodology on synthetic data and validate its effectiveness through an application to time-dependent crime patterns across multiple U.S. cities. Supplementary files for this article are available online.
Group lasso; Matrix-variate data; Model-based clustering; Penalized likelihood; Sparse estimation;
Settore STAT-01/A - Statistica
2024
9-nov-2024
Article (author)
File in questo prodotto:
File Dimensione Formato  
JCGS-CappozzoCasaFop_sparsemixmat_unblided.pdf

accesso aperto

Tipologia: Pre-print (manoscritto inviato all'editore)
Dimensione 787.08 kB
Formato Adobe PDF
787.08 kB Adobe PDF Visualizza/Apri
Sparse model-based clustering of three-way data via lasso-type penalties.pdf

embargo fino al 09/11/2025

Tipologia: Post-print, accepted manuscript ecc. (versione accettata dall'editore)
Dimensione 2.41 MB
Formato Adobe PDF
2.41 MB Adobe PDF   Visualizza/Apri   Richiedi una copia
Pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/2434/1118708
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 0
  • ???jsp.display-item.citation.isi??? 0
social impact