Finite Gaussian mixture models provide a powerful and widely employed probabilistic approach for clustering multivariate continuous data. However, the practical usefulness of these models is jeopardized in high-dimensional spaces, where they tend to be over-parameterized. As a consequence, different solutions have been proposed, often relying on matrix decompositions or variable selection strategies. Recently, a methodological link between Gaussian graphical models and finite mixtures has been established, paving the way for penalized model-based clustering in the presence of large precision matrices. Notwithstanding, current methodologies implicitly assume similar levels of sparsity across the classes, not accounting for different degrees of association between the variables across groups. We overcome this limitation by deriving group-wise penalty factors, which automatically enforce under or over-connectivity in the estimated graphs. The approach is entirely data-driven and does not require additional hyper-parameter specification. Analyses on synthetic and real data showcase the validity of our proposal

(2022). Group‐Wise Shrinkage Estimation in Penalized Model‐Based Clustering [journal article - articolo]. In JOURNAL OF CLASSIFICATION. Retrieved from https://hdl.handle.net/10446/269559

Group‐Wise Shrinkage Estimation in Penalized Model‐Based Clustering

Casa, Alessandro;
2022-01-01

Abstract

Finite Gaussian mixture models provide a powerful and widely employed probabilistic approach for clustering multivariate continuous data. However, the practical usefulness of these models is jeopardized in high-dimensional spaces, where they tend to be over-parameterized. As a consequence, different solutions have been proposed, often relying on matrix decompositions or variable selection strategies. Recently, a methodological link between Gaussian graphical models and finite mixtures has been established, paving the way for penalized model-based clustering in the presence of large precision matrices. Notwithstanding, current methodologies implicitly assume similar levels of sparsity across the classes, not accounting for different degrees of association between the variables across groups. We overcome this limitation by deriving group-wise penalty factors, which automatically enforce under or over-connectivity in the estimated graphs. The approach is entirely data-driven and does not require additional hyper-parameter specification. Analyses on synthetic and real data showcase the validity of our proposal
articolo
2022
Casa, Alessandro; Cappozzo, Andrea; Fop, Michael
(2022). Group‐Wise Shrinkage Estimation in Penalized Model‐Based Clustering [journal article - articolo]. In JOURNAL OF CLASSIFICATION. Retrieved from https://hdl.handle.net/10446/269559
File allegato/i alla scheda:
File Dimensione del file Formato  
s00357-022-09421-z.pdf

accesso aperto

Versione: publisher's version - versione editoriale
Licenza: Creative commons
Dimensione del file 2.05 MB
Formato Adobe PDF
2.05 MB Adobe PDF Visualizza/Apri
Pubblicazioni consigliate

Aisberg ©2008 Servizi bibliotecari, Università degli studi di Bergamo | Terms of use/Condizioni di utilizzo

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/10446/269559
Citazioni
  • Scopus 2
  • ???jsp.display-item.citation.isi??? 2
social impact