Data mining based reduction on credit evaluation index of bank personal customer

摘要

Making a credit evaluation to bank personal customer is an essential way to eliminate the risk for banks. But most credit evaluation index systems of bank customers in most researches are over complicated and difficult to apply. The paper attempts to accomplish reduction analysis on credit evaluation index system of bank customer based on data mining method, including cluster method and decision tree method. In this paper, a common credit evaluation index system of bank customer is constructed. Moreover, the rationality of the index system is analyzed by clustering method, and according to which, the index system is reduced to 8 indexes from 18 indexes. Furthermore, the efficiency of the reduced index system is verified by decision tree method. The reduced credit evaluation index system is more efficient with the analysis cost declined.