par Chen, Dan;Devillers, Olivier;Iacono, John J.;Langerman, Stefan ;Morin, Pat P.
Référence Computational geometry, 46, 2, page (140-147)
Publication Publié, 2013-02
Article révisé par les pairs
Résumé : Oja depth (Oja 1983) is a generalization of the median to multivariate data that measures the centrality of a point x with respect to a set S of points in such a way that points with smaller Oja depth are more central with respect to S. Two relationships involving Oja depth and centers of mass are presented. The first is a form of Centerpoint Theorem which shows that the center of mass of the convex hull of a point set has low Oja depth. The second is an approximation result which shows that the center of mass of a point set approximates a point of minimum Oja depth. © 2012 Elsevier B.V. All rights reserved.