Skip to Main Content
According to analysis and practical situation of credit risk assessment in commercial banks, some indexes are selected to establish the index system. The credit risk classes are separated into two classes- normality and loss. To classify the credit risk data, support vector machines (SVM) model based on PCA (principal component analysis) is established. In order to verify the effectiveness of the method, a real case is given and SVM model without using PCA is also used to classify the same data. The experimental results show that SVM model based on PCA is effective in credit risk assessment and achieves better performance than SVM model without using PCA.