A penalized criterion for variable selection in classification - Archive ouverte HAL Accéder directement au contenu
Article Dans Une Revue Journal of Multivariate Analysis Année : 2007

A penalized criterion for variable selection in classification

Résumé

In this paper, the problem of variable selection in classification is considered. On the basis of recent developments in model selection theory, we provide a criterion based on penalized empirical risk, where the penalization explicitly takes into account the number of variables of the considered models. Moreover, we give an oracle-type inequality that non-asymptotically guarantees the performance of the resulting classification rule. We discuss the optimality of the proposed criterion and present an application of the main result to backward and forward selection procedures.

Dates et versions

hal-01197536 , version 1 (11-09-2015)

Identifiants

Citer

Tristan Mary-Huard, Stephane Robin, Jean-Jacques Daudin. A penalized criterion for variable selection in classification. Journal of Multivariate Analysis, 2007, 98 (4), pp.695-705. ⟨10.1016/j.jmva.2006.06.003⟩. ⟨hal-01197536⟩
35 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More