A Least Squares Bilateral-Weighted Fuzzy SVM Method to Evaluate Credit Risk

In this study, we propose a least squares bilateral-weighted fuzzy support vector machine (LS-BFSVM) method to evaluate the credit risk problem. The method can not only reduce the computational complexity by considering equality constraints instead of inequalities for the classification problem with a formulation in least squares sense, but also increase the training algorithm's generalization ability by treating each training sample as being both a possible good and bad customer and considering bilateral-weighted classification errors. For illustration purpose, a real-world credit risk assessment dataset is used to test the effectiveness of the LS-BFSVM method.