A DC programming approach for feature selection in the Minimax Probability Machine

DOI

10.1080/18756891.2013.864471How to use a DOI?

Keywords

Feature selection, Minimax probability machine, DC programming

Abstract

This paper presents a new feature selection framework based on the -norm, in which data are summarized by their moments of the class conditional densities. However, discontinuity of the -norm makes it difficult to find the optimal solution. We apply a proper approximation of the -norm and a bound on the misclassification probability involving the mean and covariance of the dataset, to derive a robust difference of convex functions (DC) program formulation, while the DC optimization algorithm is used to solve the problem effectively. Furthermore, a kernelized version of this problem is also presented in this work. Experimental results on both real and synthetic datasets show that the proposed formulations can select fewer features than the traditional Minimax Probability Machine and the -norm state.

Copyright

© 2017, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - JOUR
AU  - Liming Yang
AU  - Ribo Ju
PY  - 2014
DA  - 2014/02/03
TI  - A DC programming approach for feature selection in the Minimax Probability Machine
JO  - International Journal of Computational Intelligence Systems
SP  - 12
EP  - 24
VL  - 7
IS  - 1
SN  - 1875-6883
UR  - https://doi.org/10.1080/18756891.2013.864471
DO  - 10.1080/18756891.2013.864471
ID  - Yang2014
ER  -