Optimal Expected-Distance Separating Halfspace

E. Carrizosa, Frank Plastria

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

One recently proposed criterion to separate two data sets in discriminant analysis is to use a hyperplane, which minimizes the sum of distances to it from all the misclassified data points. Here all distances are supposed to be measured by way of some fixed norm, while misclassification means lying in the wrong halfspace. In this paper we study the problem of determining such an optimal halfspace when points are distributed according to an arbitrary random vector X in Rd. In the unconstrained case in dimension d, we prove that any optimal separating halfspace always balances the misclassified points. Moreover, under polyhedrality assumptions on the support of X, there always exists an optimal separating halfspace passing through d affinely independent points. These results extend in a natural way when different norms (or a fixed gauge) are used to measure distances, and we allow constraints modeling that certain points are forced to be correctly classified.
Original languageEnglish
Pages (from-to)662-677
Number of pages16
JournalMathematics of Operations Research
Volume33
Issue number3
Publication statusPublished - Aug 2008

Bibliographical note

Working paper MOSI/7, Vakgroep MOSI, VUB, maart 2004, 26p.

Keywords

  • gauge-distance to hyperplane
  • separating halfspace
  • discriminant analysis

Fingerprint Dive into the research topics of 'Optimal Expected-Distance Separating Halfspace'. Together they form a unique fingerprint.

Cite this