The Latent Block Model: a useful model for high dimensional data

Christine Keribin 1, 2 Gilles Celeux 1 Valérie Robert 1, 2
1 SELECT - Model selection in statistical learning
Inria Saclay - Ile de France, LMO - Laboratoire de Mathématiques d'Orsay, CNRS - Centre National de la Recherche Scientifique : UMR
Abstract : The Latent Block Model (LBM) designs in a same exercise a clustering of the rows and the columns of a data array. Typically the LBM is expected to be useful to analyze huge data sets with many observations and many variables. But it encounters several numerical issues with big data set: maximum likelihood is jeopardized by spurious maxima and selecting a proper model is challenging since there are a lot of models are in competition. In this communication, we analyze these numerical issues. In particular, we make use of Bayesian inference to avoid spurious solutions and propose an efficient way to scan the model set. Moreover, we advocate the exact Integrated Completed Likelihood (ICL) criterion to select a proper and consistent LBM. The methods and algorithms will be ilustrated with pharmacovigilance data involving large arrays of data.
Type de document :
Communication dans un congrès
ISI 2017 - 61st world statistics congress, Jul 2017, Marrakech, Morocco. pp.1-6
Liste complète des métadonnées

Littérature citée [13 références]  Voir  Masquer  Télécharger

https://hal.inria.fr/hal-01658589
Contributeur : Christine Keribin <>
Soumis le : jeudi 7 décembre 2017 - 17:11:07
Dernière modification le : lundi 15 janvier 2018 - 08:10:02

Fichier

KERIBIN-CELEUX-ROBERT-ISI17.pd...
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-01658589, version 1

Citation

Christine Keribin, Gilles Celeux, Valérie Robert. The Latent Block Model: a useful model for high dimensional data. ISI 2017 - 61st world statistics congress, Jul 2017, Marrakech, Morocco. pp.1-6. 〈hal-01658589〉

Partager

Métriques

Consultations de la notice

231

Téléchargements de fichiers

52