Data mining-based credit evaluation for users of credit card

As the most important tasks of a bank, assessment of credit card users is aimed to keep the risk of a credit loss low and to minimize costs of failure over risk groups. Typical method requests applicants to fill a form and give each item a suitable score according to a predefined scoring table. The sum will direct a bank to determine whether the applicant is accepted. Each item´s score in the scoring table is given by experts, as something of experience it is subjective and the tightness of it is unknown. Additionally, if condition changes, adjusting the scoring table is difficult. A data mining-based approach IGCSM is introduced to solve the problem of assessing credit card applicants. It works on personal information and consumptive data. Firstly, a decision tree is constructed by ID3 algorithm. Then the tree plus the information gain of each non-leaf node are used to give an objective estimation of each attribute´s classification contribution i.e. score. Experiment on real-life data shows that this method has higher correctness than the typical method and can be modified automatically when condition changes.