Credit Risk Assessment Based on Fuzzy SVM and Principal Component Analysis

Credit risk assessment has been an important research topic in customer relationship management. It is also an important field for commercial banks because discriminating good creditors from bad ones is becoming more and more crucial for banks. A Fuzzy Support Vector Machine (FSVM) classification model based on principal component analysis (PCA-FSVM) was advanced, which adapted PCA to extract principal components to replace the original indexes, so that the processing speed and classification accuracy can be improved. Then credit risk assessment example that apply this classification model was provided and compared with the method of SVM and BP neural networks, which shows the better performance and better classification accuracy of PCA-FSVM.