Identifying the Impact of Fraud on Corporate Customers' Credit Scoring by Data Mining Approaches

Abstract

Credit risk is one of the most important risks which banks and financial organizations face. It is related to unpaid and delayed installments. Banks evaluate their customers' credit in order to prevent this hazard. Development banks, which are the focus of this research, fund facilities based on working capital, so customers sometimes do fraud in declaring working capital. Considering fraud consequences and making a credit scoring model with sensitivity to fraud are the main aims of this research. The statistical population of this research includes companies who have referred to branches of an Iranian Bank. This research includes 55 financial and non-financial variables based on the credit scoring model. In the first step, fraudulent companies have been realized. Finally, in order to offer an optimized and sustainable model through merging machine learning methods and reporting performance evaluation indicators, the impacts of fraud have been considered.