Rough Margin-Based Linear v Support Vector Machine

Abstract

Rough set theory is introduced into linear νsupport vector machine (svm), and rough margin-based linear νsvm is proposed in this paper. By constructing rough lower margin, rough upper margin and rough boundary in linear ν svm, then we maximize the rough margin not margin in linear ν svm. Thus more points are considered in constructing the separating hyper-plane than those used in linear υ svm. Moreover, different points in different positions are proposed to have different effect on the separating hyper-plane, where points in the lower margin have more effects than those in the boundary of the rough margin. The proposed algorithm is compared with other svm algorithms, the experiment results demonstrate the feasibility and validity of the proposed algorithm.