Developing a model for validation and prediction of bank customer credit using information technology (case study of Dey Bank)

Abstract

Credit risk is the most important risk of banks. The main approaches of the bank to reduce credit risk are correct validation using the final status and the validation model parameters. High fuel of bank reserves and lost or outstanding facilities of banks indicate the lack of appropriate validation models in the banking network. The weakness of the previous models is due to the choice of inappropriate decision parameters, technical weakness of the model and lack of access to desired data. In this paper, in order to establish a communication between the final status and the parameters of facilities granted, data mining technique with the help of machine learning and neural networks have been used. A database of facilities granted by Dey Bank was created and a model with data mining approach was prepared. This model has good accuracy is able to validate real customer. According to the analysis, interest rate parameter is more important in determining a customer validation. The model has higher accuracy and comprehensiveness compared to the similar cases, due to the database size, type of data mining and learning algorithms applied.