An entropy criterion for assessing the number of clusters in a mixture model

DOI10.1007/BF01246098zbMath0861.62051MaRDI QIDQ1126397

Gilles Celeux, Gilda Soromenho

Publication date: 27 April 1997

Published in: Journal of Classification (Search for Journal in Brave)

zbMATH Keywords

likelihood; entropy; Monte Carlo; Gaussian mixture; cluster analysis; mixture model; number of clusters; entropy criterion; Bayesian criteria; classification likelihood

Mathematics Subject Classification ID

62H30: Classification and discrimination; cluster analysis (statistical aspects)

62B10: Statistical aspects of information-theoretic topics

Related Items

Multilevel Functional Clustering Analysis, Choosing models in model-based clustering and discriminant analysis, Correlating Two Continuous Variables Subject to Detection Limits in the Context of Mixture Distributions, Determining the number of components in mixtures of linear models., Mixture model clustering for mixed data with missing information, Gaussian mixture modelling to detect random walks in capital markets, The generalized cross entropy method, with applications to probability density estimation, Optimization of the number of components in the mixed model using multi-criteria decision-making, Order selection in finite mixtures of linear regressions, Dissolution point and isolation robustness: Robustness criteria for general cluster analysis methods, Fitting of mixtures with unspecified number of components using cross validation distance estimate, Clustering objects described by juxtaposition of binary data tables, Model-based cluster and discriminant analysis with the MIXMOD software, Gamma mixture models for target recognition, Breakdown points for maximum likelihood estimators of location-scale mixtures, A person-fit index for polytomous Rasch models, latent class models, and their mixture generalizations, Latent Markov model for longitudinal binary data: an application to the performance evaluation of nursing homes, Finite mixtures of multivariate Poisson distributions with application, Describing disability through individual-level mixture models for multivariate binary data, A unified view on clustering binary data, Density estimation and comparison with a penalized mixture approach, Finding standard dental arch forms from a nationwide standard occlusion study using a Gaussian functional mixture model, Assessment of the number of components in Gaussian mixture models in the presence of multiple local maximizers, Normality-based validation for crisp clustering, Model-based linear clustering, Testing the number of components of the mixture of two inverse Weibull distributions

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Multi-sample cluster analysis using Akaike's information criterion
• Another interpretation of the EM algorithm for mixture distributions
• On some significance tests in cluster analysis
• Large-sample results for optimization-based clustering methods
• Clustering criteria for discrete data and latent class models
• Estimating the dimension of a model
• Comparing approaches for testing the number of components in a finite mixture model
• On the information-based measure of covariance complexity and its application to the evaluation of multivariate linear models
• Classification and Mixture Approaches to Clustering via Maximum Likelihood
• Mixture Models, Outliers, and the EM Algorithm
• 389: Separating Mixtures of Normal Distributions
• Asymptotic behaviour of classification maximum likelihood estimates
• Model-Based Gaussian and Non-Gaussian Clustering
• A new look at the statistical model identification