Linear lJ-nonparallel support vector machine for pattern classification

Abstract

In this paper, we propose a novel nonparallel linear hyperplane classifier called linear $\nu $-nonparallel support vector machine ($ L_{1}-\nu $-NPSVM) for binary classification. Based on $L_{1}-$ NPSVM (Linear Nonparallel Support Vector Machine), and combining the $\nu $-support vector classification and $\nu $-support vector regression together, the primal problem of $ L_{1}-\nu $-NPSVM is obtained. Compared to $L_{1}$-NPSVM, $ L_{1}-\nu $-NPSVM has the following advantages: (1) By introducing a new parameter $\nu $ to effectively control the number of support vectors, the model's generalization ability and accuracy can be improved; (2) By introducing a new parameter v, we can eliminate one of the other free parameters of the $L_{1}$-NPSVM to reduce the difficulty of selecting parameters. Moreover, experimental results on data sets show the effectiveness of our method.