Fuzzy neural network model for assessing credit risk in commercial banks

Abstract

An adaptive fuzzy neural network has been developed to extract fuzzy rules directly from data sample for credit risks assessment. As an application example, a model has been constructed to assess the credit risks in commercial banks by using adaptive neuro-fuzzy inference method. Numerical example indicates fuzzy neural network could fit to assess the credit risk in commercial banks. Furthermore, the results also indicate that the prediction based on fuzzy neural network model is accurate.