[VS08] Clustering and Disjoint Principal Component Analysis

Revue Internationale avec comité de lecture : Journal Computational Statistics & Data Analysis, vol. 53(8), pp. 3194-3208, 2008

Mots clés: PCA, cluster analysis

Résumé: A constrained principal component analysis, which aims at a simultaneous clustering of objects and a partitioning of variables is proposed. The new methodology allows to identify components with maximum variance, each one a linear combination of a subset of variables. All the subsets form a partition of variables. Simultaneously, a partition of objects is also computed maximizing the between cluster variance. The methodology is formulated in a semi-parametric least-squares framework as a quadratic mixed continuous and integer problem. An alternating leastsquares algorithm is proposed to solve the clustering and disjoint PCA. Two applications are given to show the features of the methodology.

Equipe: msdma


@article {
title="{Clustering and Disjoint Principal Component Analysis}",
author="M. Vichi and G. Saporta",
journal="Computational Statistics & Data Analysis",