The Application Of Hybrid-Kernel Function Support Vector Machine In Credit Evaluation

With China's rapid economy development and in condition of continuously heated consumer credit, commercial banks take consumer loans as an important component of their future development strategy. However currently domestic commercial banks' level of risk management on consumer loans are low. In consumer credit payment process, they still use traditional methods of credit analysis to evaluate the consumer credit applicants' credit standing and their ability to repay. There has been no reliable model in personal credit scoring methods. Although in current academic community, kinds of credit scoring model are presented including mathematical statistics, artificial intelligence, these are impossible to achieve optimum state of both accuracy and stability.Based on the analysis about the development history of personal credit scoring and the application status of personal credit scoring models, a newly emergence support vector machine theory has been applied to develop personal credit evaluation model and take advantage of practical personal credit data of a commercial bank to do practical research. Firstly, the paper constructs a personal credit evaluation system, which suits our country's current situation. Secondly, we analyze factors that influence performance of support vector macine and problems exist in standard support vector machine. Aiming to solve these problems, a scheme of parameter optimization of support vector machine and a hybrid kernel, which combines the advantages of single kernel function, has been designed. In the final analysis, a comparison concering accuracy and steadiness among hybridy kernel function support vector machine, classical statistics model and neural network has been made.