The parameters of support vector machine (SVM) are crucial to the model's classification performance.  PSO-SVM model was constructed by using particle swarm optimization (PSO) to search for the optimal parameters of SVM, after the randomness of parameters selection in SVM was considered.The modified PSO was used to optimize the parameters of SVM and the particle's fitness function was carried out to control the type II error rate in personal credit scoring，which costs great loss to commercial banks.   The results indicate that PSO-SVM model gets high classification accuracy with low type II error rate and the model shows a strong robustness, which presents more applicable for commercial banks to control the consumer credit risks.