Local Subspace Classifier in Reproducing Kernel Hilbert Space

  • Authors:
  • Dongfang Zou

  • Affiliations:
  • -

  • Venue:
  • ICMI '00 Proceedings of the Third International Conference on Advances in Multimodal Interfaces
  • Year:
  • 2000

Quantified Score

Hi-index 0.00

Visualization

Abstract

Local Subspace Classifier (LSC) is a new classification technique, which is closely related to the subspace classification methods, and a heir of prototype classification methods. And it is superior to both of them. In this paper, a method of improving the performance of Local Subspace Classifier is presented. It is to avoid the intersection of the local subspaces representing the respective categories by mapping the original feature space into RKHS (Reproducing Kernel Hilbert Space).