With the development of Chinas economy,credit scoring has become important.The general credit scoring model is to solve the two classification problems,but in real life we often encounter multiple classification problems.This paper proposes a multi-class support vector machine based on genetic algorithm,which can solve multiple classification problems in the behavior assessment model.Genetic algorithm is used for support vector machines parameters optimization search.The parameters of SVM are coded into chromosomes with Gray code strategy.The empirical results show that the algorithm is very practical,and it has good prediction accuracy.