HAL : hal-00492406, version 4

DOI : 10.1007/s11222-011-9249-9

Fiche détaillée

Récupérer au format BibTeX - EndNote - TEI - RefWorks -

Statistics and Computing 22, 1 (2012) 301--324

Versions disponibles :

v1 (15-06-2010)

v2 (28-09-2010)

v3 (12-01-2011)

v4 (19-04-2011)

Simultaneous model-based clustering and visualization in the Fisher discriminative subspace

Charles Bouveyron 1, Camille Brunet 2

(2012)

Clustering in high-dimensional spaces is nowadays a recurrent problem in many scientific domains but remains a difficult task from both the clustering accuracy and the result understanding points of view. This paper presents a discriminative latent mixture (DLM) model which fits the data in a latent orthonormal discriminative subspace with an intrinsic dimension lower than the dimension of the original space. By constraining model parameters within and between groups, a family of 12 parsimonious DLM models is exhibited which allows to fit onto various situations. An estimation algorithm, called the Fisher-EM algorithm, is also proposed for estimating both the mixture parameters and the discriminative subspace. Experiments on simulated and real datasets show that the proposed approach performs better than existing clustering methods while providing a useful representation of the clustered data. The method is as well applied to the clustering of mass spectrometry data.

1 :	Statistique, Analyse et Modélisation Multidisciplinaire (SAmos-Marin Mersenne) (SAMM)
	Université Paris I - Panthéon-Sorbonne
2 :	Informatique, Biologie Intégrative et Systèmes Complexes (IBISC)
	Université d'Evry-Val d'Essonne : EA4526

Équipe de recherche : TADIB

Domaine

:

Mathématiques/Statistiques Statistiques/Théorie

Mots Clés : High-dimensional clustering – Model-based clustering – Discriminative subspace – Fisher criterion – Visualization – Parsimonious models

hal-00492406, version 4

http://hal-paris1.archives-ouvertes.fr/hal-00492406

oai:hal-paris1.archives-ouvertes.fr:hal-00492406

Contributeur : Charles Bouveyron <>

Soumis le : Mardi 19 Avril 2011, 10:16:21

Dernière modification le : Mardi 10 Janvier 2012, 11:55:58