Neural Computing and Applications

, Volume 23, Issue 7, pp 2159–2166

Mixed-norm linear support vector machine

  • Chunhua Zhang
  • Yuanhai Shao
  • Junyan Tan
  • Naiyang Deng
Original Article

DOI: 10.1007/s00521-012-1166-0

Cite this article as:
Zhang, C., Shao, Y., Tan, J. et al. Neural Comput & Applic (2013) 23: 2159. doi:10.1007/s00521-012-1166-0

Abstract

This paper presents a new version of support vector machine (SVM) named l2 − lp SVM (0 < p < 1) which introduces the lp-norm (0 < p < 1) of the normal vector of the decision plane in the standard linear SVM. To solve the nonconvex optimization problem in our model, an efficient algorithm is proposed using the constrained concave–convex procedure. Experiments with artificial data and real data demonstrate that our method is more effective than some popular methods in selecting relevant features and improving classification accuracy.

Keywords

Support vector machineOptimizationNormFeature selectionConstrained concave–convex procedure

Copyright information

© Springer-Verlag London Limited 2012

Authors and Affiliations

  • Chunhua Zhang
    • 1
  • Yuanhai Shao
    • 2
  • Junyan Tan
    • 3
  • Naiyang Deng
    • 3
  1. 1.Department of Mathematics, Information SchoolRenmin University of ChinaBeijingChina
  2. 2.Zhijiang CollegeZhejiang University of TechnologyHangzhouChina
  3. 3.College of ScienceChina Agricultural UniversityBeijingChina