Searching parameter values in support vector machines using DNA genetic algorithms

Abstract

A novel DNA encoding genetic algorithm, called SVM-DNAGA, is proposed to search for optimal values for the parameters in support vector machines. With this algorithm, the training process of support vector machines can converge quickly and the performance of the support vector machines can improve. The parameters in the support vector machines are encoded into chromosomes using DNA encoding. DNA genetic operations, including selection, transgenosis and frameshift mutation, are used in SVM-DNAGA. Four datasets are used in the computational experiments to verify the effectiveness of SVM-DNAGA. Compared with other commonly used classifiers, SVM-DNAGA obtains very good results.