Fuzzy Support Vector Machine Based on Vague Sets for Credit Assessment

Credit assessment is one of important tools help financial institutions to hedge the credit risks. Support vector machine and fuzzy support vector machine have been applied in Credit assessment field. Fuzzy support vector machine (FSVM) does not treat equally the input point. It improves the generalization power of traditional SVM by applying a fuzzy membership to each input data point. In this paper, we propose a new FSVM based on vague sets that apply a truth-membership and a false- membership to each data point of training sets. In order to verity the effectiveness of the new FSVM, a real case of home loan data sets is given and the experimental results show that the model is promising.