Research of Default Risk of Commercial Bank´s Personal Loan Based on Rough Sets and Neural Network

Guard against financial risks, reduce bad loans, increase the ability to identity risk of commercial banks, the key is risk warning. In view of the increasing proportion of personal loans in banking business, it is particularly important to warning personal loans default risk. Commercial bank lending itself is a complex nonlinear system, using general linear theory is difficult to objectively reflect the laws of this, this paper uses rough set and BP neural network. Personal loan default index first constructed, and then use rough sets to streamline, and then BP neural network was trained on the samples to determine risk of default. Results showed that rough set and BP neural network test samples of prediction accuracy.