New multi-class classification method based on the SVDD model

  • Authors:
  • Lei Yang;Wei-Min Ma;Bo Tian

  • Affiliations:
  • Shanghai Key Laboratory of Power Station Automation Technology, School of Mechatronics Engineering and Automation, Shanghai University, Shanghai, China;School of Economics and Management, Tongji University, Shanghai, China;School of Information Management and Engineering, Shanghai University of Finance Economic, Shanghai, China

  • Venue:
  • ISNN'11 Proceedings of the 8th international conference on Advances in neural networks - Volume Part II
  • Year:
  • 2011

Quantified Score

Hi-index 0.00

Visualization

Abstract

New decision-making function for multi-class support vector domain description (SVDD) classifier using the conception of attraction force was proposed in this paper. As for multi-class classification problems, multiple optimized hyperspheres which described each class of dataset were constructed separately similar with in the preliminary SVDD. Then new decision-making function was proposed based on the parameters of the multi-class SVDD model with the conception of attraction force. Experimental results showed that the proposed decision-making function for multi-class SVDD classifier is more accurate than the decision-making function using local density degree.