Credit Risk Evaluation System: An Artificial Neural Network Approach
Decisions concerning credits granting are one of the most crucial issues in an everyday banks’ policy. Well-allocated credits may become one of the biggest sources of profits for any financial organizations. On the other hand, this kind of bank’s activity is connected with high risk as big amount of bad decisions may even cause bankruptcy. The key problem consists of distinguishing good (that surely repay) and bad (that likely default) credit applicants. Credit risk evaluation is an important and interesting management science problem in financial analysis. The main idea in credit risk evaluation investigations consists of building classification rules that properly assess bank customers as good or bad. In this paper, we proposed an architecture which uses the theory of artificial neural networks and business rules to correctly determine whether a customer is good or bad. In the first step, by using clustering algorithm, clients are segmented into groups with similar features. In the second step, decision trees are built based on classification rules defined for each group of clients. To avoid redundancy, different attributes are taken into consideration during each phase of classification. The proposed approach allows for using different rules within the same data set, and for defining more accurately clients with high risk. Preliminary result indicates that the model presented is promising and reasonable.