Research On Integrated Model Of Personal Credit Assessment

With the trend of financial globalization and financial market volatility increased, national banks and investors face more and more serious financial risks. The World Bank has shown that the main reason leading to bank failures is the credit risk.At present,many scholars internal or abroad have been extensively studied in personal credit evaluation,designed many models meeting special financial environment. The review undertaken in this paper covers credit evaluation and ensemble methods. At first, this paper discusses the conception and features of credit risk,as well as the conception of personal credit assessment.Then it introduces some models of credit risk evaluation from two aspects of traditional statistical methods and modern artificial intelligence methods. On this basis, the integration models of Bagging classifier and the single classifier is applied to the two group database. The database is used to test the accurability and stability of the models.The results shows Bagging classification model has higher classifition accuracy compared to single classifier.The proposed Bagging-Bagging model is compared with the Bagging integrated model,the result shows that Bagging-Bagging has better classification effect,can be used in personal credit evaluation innovatively.Therefore,we can think,Bagging Ensemble model is more suitable in the field of personal credit assessment.lt can improve the classication effect,especially compared to models which have weak classication performance.