Support Vector Machine with feature selection: A multiobjective approach

Support Vector Machines are models widely used in supervised classification. The classical model minimizes a compromise between the structural risk and empirical risk. In this paper, we consider Machine with feature selection design implement bi-objective evolutionary algorithm for approximating Pareto optimal frontier of two objectives. metaheuristic is based on non-dominated sorting genetic includes problem-specific knowledge. To demonstrate efficiency proposed, have carried out extensive computational experiments comparing Pareto-frontiers given by exact method AUGMECON2 approach respectively set well known instances. also discuss some properties points frontier.