Multi-objective optimization for support vector regression parameters

The selection of model's parameters and the designing sparse model is the key step for the calculate efficiency, generalization performance and calculate efficiency of support vector regression(SVR) model. A new parameter setting methods for building SVR model was proposed based on multi-object optimal algorithms. Aiming at improving the regression accuracy and generalization performance of the model, non-dominated sorting genetic algorithm (NSGA-II) was adopted to get the optimal parameters from the Pareto optimal solution set by solving multi-objective problems. Simulation results show that the model constructed by the optimal parameters has superior learning accuracy and generalization performance.