Design Of Credit Card Risk Management System On CHAID Decision Tree

Owing to the lack of experience in Credit card business, Xinrui Bank is relatively weak in credit risk management. Currently review process of card issuing only depends on credit information data from PBOC system, which isn’t enough. And with the development of credit card business, a massive historical transaction data will be generated, so effective management and analysis of these data is needed. In order to avoid credit card fraud and reduce credit card bad debt risk, on the one hand analyzing the customers’ basic information, which controls credit card risk from the source. On the other hand, some unknown and valuable rules can be found out from the massive credit card transaction data by means of using CHAID decision tree. Then prediction can be achieved by these rules, which undoubtedly reduces the commercial bank credit card malicious overdraft and other risks.In the beginning current situation of Xinrui bank as well as the lack of the response management of various risks is discussed. It is very necessary to reduce the credit risk through finding useful information from data analysis with the method of CHAID decision tree. On the basis of clear system design, main functions such as the credit card customer management, card applications and consumption authorization are discussed, and core processes such as customer credit scoring and risks warning also are been analyzed. Then make detailed design for the core subsystems like the customer credit management, card issuing audit and credit fraud. It is focused on building credit scoring to get credit rating classification rules and using CHAID decision tree to recognize credit fraud process, to approve credit card issuing transactions and consumer behaviors predictable application. At last, the system is compared with the similar risk management system.