Multiple Kernel Fuzzy Support Vector Machine Based on Kernel-Target Alignment 

Jiaoyang Zhang, Qiang He

School of Science, Beijing University of Civil Engineering and Architecture, Beijing, 100044, China

Abstract: Support vector machines (SVMs) are currently widely used machine learning techniques. SVM is used to construct an optimal hyper-plane that implies an extraordinary generalization capability and good performances. So far, SVMs have already been successfully applied to many real fields. In view of the difficulties in kernel selection and sensitivity to noise, we propose Kernel-Target alignment based multiple kernel fuzzy support vector machine in this paper. The Kernel-Target alignment based multiple kernel learning is introduced to multiple kernel fuzzy support vector machine. It not only avoids the problem of kernel selection, but also improves the robustness to noise. The experimental simulation also validates the feasibility and effectiveness of the method.

Keywords: Kernel-Target alignment; Multiple kernel learning; Support vector machine; Fuzzy