Application Of Data Mining In The Commercial Banks’ Credit Decision-making Process

The interest spreads of deposit and loan is an important way of bank profit. Loan asset quality directly threatens the existence of the bank. When banks are being in pursuit of maximum profits, they also have to face the great risk. How to avoid the potential risk is an important issue that bank’s credit sector has to face. In order to guard against all kinds of credit risk better, banks must establish a perfect and effective scientific credit rating model, to control the quality of loans from the source.It can help banks to create maximum value that establishing intellective credit management system, in which the bank’s various business functions are integrated, fully exploiting all kinds of business data of banks. This paper mainly related to the data of a certain bank integrated business system, credit management system and credit system as the basis, combined with the characteristics of bank data, through data sampling, data preprocessing procedure, converted the results into the data warehouse of data mining server, used logistic regression to establish a credit rating model to assess the credit risk of the borrower when he applies for a loan, and decide whether give him a loan. This paper also made an empirical analysis on the credit rating model.Finally, after the analysis of the characteristics of the traditional credit system, according to the bank operation mode and the design method of decision support system, this paper established a bank credit decision support system based on the logistic regression, form which some mining information beneficial to bank credit decision could be got. This information will provide scientific reference for bank credit decision-making.