How useful Bayesian inference could be in Model-based clustering? - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2013

How useful Bayesian inference could be in Model-based clustering?

Gilles Celeux

Abstract

In this communication, we analyse the pro and the con of Bayesian inference in the model-based clustering context. We exhibit situations where its main drawbacks can be avoided or circumvented. We consider the latent class model for categorical data and derive their (completed) integrated likelihoods without requiring asymptotic approximations. We highlight the interest and the traps of the resulting model selection criteria.
No file

Dates and versions

hal-00927006 , version 1 (10-01-2014)

Identifiers

  • HAL Id : hal-00927006 , version 1

Cite

Gilles Celeux. How useful Bayesian inference could be in Model-based clustering?. Advances in Latent Variables-Methods, Models and Applications, SOCIETÀ ITALIANA DI STATISTICA, Jun 2013, Brescia, Italy. ⟨hal-00927006⟩
78 View
0 Download

Share

Gmail Facebook X LinkedIn More