LP-SVR Model Selection Using an Inexact Globalized Quasi-Newton Strategy

HTML  XML Download Download as PDF (Size: 460KB)  PP. 19-28  
DOI: 10.4236/jilsa.2013.51003    5,492 Downloads   7,771 Views  Citations

ABSTRACT

In this paper we study the problem of model selection for a linear programming-based support vector machine for regression. We propose generalized method that is based on a quasi-Newton method that uses a globalization strategy and an inexact computation of first order information. We explore the case of two-class, multi-class, and regression problems. Simulation results among standard datasets suggest that the algorithm achieves insignificant variability when measuring residual statistical properties.

Share and Cite:

P. Rivas-Perea, J. Cota-Ruiz, J. Venzor, D. Chaparro and J. Rosiles, "LP-SVR Model Selection Using an Inexact Globalized Quasi-Newton Strategy," Journal of Intelligent Learning Systems and Applications, Vol. 5 No. 1, 2013, pp. 19-28. doi: 10.4236/jilsa.2013.51003.

Copyright © 2024 by authors and Scientific Research Publishing Inc.

Creative Commons License

This work and the related PDF file are licensed under a Creative Commons Attribution 4.0 International License.