Neural Processing Letters

, Volume 9, Issue 3, pp 293–300

Least Squares Support Vector Machine Classifiers

  • J.A.K. Suykens
  • J. Vandewalle
Article

DOI: 10.1023/A:1018628609742

Cite this article as:
Suykens, J. & Vandewalle, J. Neural Processing Letters (1999) 9: 293. doi:10.1023/A:1018628609742
  • 12k Views

Abstract

In this letter we discuss a least squares version for support vector machine (SVM) classifiers. Due to equality type constraints in the formulation, the solution follows from solving a set of linear equations, instead of quadratic programming for classical SVM's. The approach is illustrated on a two-spiral benchmark classification problem.

classificationsupport vector machineslinear least squaresradial basis function kernel

Copyright information

© Kluwer Academic Publishers 1999

Authors and Affiliations

  • J.A.K. Suykens
    • 1
  • J. Vandewalle
    • 1
  1. 1.Department of Electrical EngineeringKatholieke Universiteit LeuvenLeuven (Heverlee)Belgium, e-mail