Feature selection and parameter optimization for support vector machines: A new approach based on genetic algorithm with feature chromosomes

► A genetic algorithm with feature chromosomes (GAFC) is proposed. ► The asymptotic behaviors of support vector machines (SVM) are fused with GA. ► The GAFC has not only the search ability of GA, but also has the search ability of feature chromosomes. ► The GAFC obtained good performances by optimizing feature subset and parameters of SVM simultaneously.