Parameter determination of support vector machine using scatter search approach

Support Vector Machine (SVM) is a popular data classification method with many diverse applications. SVM has many parameters, which have significant influences on the performance of SVM classifier. In this paper, a Scatter Search approach is used to find near optimal values of the SVM parameters and its kernel parameters. The proposed method integrates a scatter search approach with support vector machine using three different kernel functions, shortly (3SVM). To evaluate the performance of the proposed method, 4 benchmark datasets are used. Experiments and comparisons prove that the 3SVM is a promising approach and has a competitive performance relative to some other published methods.