Laplacian Discriminant Projection Based on Affinity Propagation

  • Authors:
  • Xueping Chang;Zhonglong Zheng

  • Affiliations:
  • Department of Computer Science, Zhejiang Normal University, Zhejiang, China;Department of Computer Science, Zhejiang Normal University, Zhejiang, China

  • Venue:
  • AICI '09 Proceedings of the International Conference on Artificial Intelligence and Computational Intelligence
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

The paper proposes a new algorithm for supervised dimensionality reduction, called Laplacian Discriminant Projection based on Affinity Propagation (APLDP). APLDP defines three scatter matrices using similarities based on representative exemplars which are found by Affinity Propagation Clustering . After linear transformation, the considered pairwise samples within the same exemplar subset and the same class are as close as possible, while those exemplars between classes are as far as possible. The experiments on several data sets demonstrate the competence of APLDP.