Babak Afshin - Department of Computer and Electrical Engineering Islamic Azad University, Qazvin, Iran
Babak Nasersharif - Electrical and Computer Engineering Department, K.N. Toosi University of Technology, Tehran, Iran

Support Vector Machine (SVM), is a powerful machine learning technique widely used for regression and classification. As a classifier, we can use SVM as a linearclassifier or kernel based classifier. In case of kernel based classification, the type of kernel function and its parametersaffect significantly on classification accuracy. In this paper, wepropose a method based on genetic algorithm to obtain a suitable kernel function based on linear combination ofconventional kernel functions. We use classification error as our genetic algorithm fitness function in order to minimize it.We evaluate the proposed approach using UCI dataset. Results show that this nonlinear combination can improve SVM true classification rate

support vector machine; pattern classification;hybrid kernel; genetic algorithm