HAL : hal-00656675, version 1

arXiv : 1201.0959

DOI : 10.1007/978-3-642-13312-1_46

Fiche détaillée

Récupérer au format BibTeX - EndNote - TEI - RefWorks -

Classification and Multivariate Analysis for Complex Data Structures, Bernard Fichet, Domenico Piccolo, Rosanna Verde and Maurizio Vichi (Ed.) (2011) 435-444

Constrained variable clustering and the best basis problem in functional data analysis

Fabrice Rossi (  Auteur correspondant ) 1, Yves Lechevallier 2

(03/2011)

Functional data analysis involves data described by regular functions rather than by a finite number of real valued variables. While some robust data analysis methods can be applied directly to the very high dimensional vectors obtained from a fine grid sampling of functional data, all methods benefit from a prior simplification of the functions that reduces the redundancy induced by the regularity. In this paper we propose to use a clustering approach that targets variables rather than individual to design a piecewise constant representation of a set of functions. The contiguity constraint induced by the functional nature of the variables allows a polynomial complexity algorithm to give the optimal solution.

1 :	Laboratoire Traitement et Communication de l'Information [Paris] (LTCI)
	Télécom ParisTech – CNRS : UMR5141
2 :	AxIS (INRIA Rocquencourt / INRIA Sophia Antipolis)
	INRIA

Domaine

:

Statistiques/Autres Informatique/Apprentissage

Mots Clés : functional data – variable clustering – best basis

hal-00656675, version 1

http://hal.archives-ouvertes.fr/hal-00656675

oai:hal.archives-ouvertes.fr:hal-00656675

Contributeur : Fabrice Rossi <>

Soumis le : Mercredi 4 Janvier 2012, 19:11:23

Dernière modification le : Mercredi 4 Janvier 2012, 19:39:41