The Community for Technology Leaders
Green Image
ABSTRACT
<p><b>Abstract</b>—We propose assessing a mixture model in a cluster analysis setting with the integrated completed likelihood. With this purpose, the observed data are assigned to unknown clusters using a maximum a posteriori operator. Then, the Integrated Completed Likelihood (ICL) is approximated using an <it>à la</it> Bayesian information criterion (BIC). Numerical experiments on simulated and real data of the resulting ICL criterion show that it performs well both for choosing a mixture model and a relevant number of clusters. In particular, ICL appears to be more robust than BIC to violation of some of the mixture model assumptions and it can select a number of clusters leading to a sensible partitioning of the data.</p>
INDEX TERMS
Mixture model, clustering, integrated likelihood, BIC, integrated completed likelihood, ICL criterion.
CITATION
Christophe Biernacki, Gérard Govaert, Gilles Celeux, "Assessing a Mixture Model for Clustering with the Integrated Completed Likelihood", IEEE Transactions on Pattern Analysis & Machine Intelligence, vol. 22, no. , pp. 719-725, July 2000, doi:10.1109/34.865189
97 ms
(Ver 3.1 (10032016))