The borrower's ability to repay the credit loan on schedule plays a major role in determining credit risk, which is the primary risk that banks today face. Today's banks are mostly exposed to credit risk, which is largely determined by the borrower's capacity to repay the credit loan on time. Credit risk evaluation includes determining the likelihood that a borrower will skip payments. Thus, there is substantial theoretical and practical value to study on bank credit risk. This paper builds an artificial neural network model to assess banks’ credit risk. This article first identifies 20 financial indicators that will be used to build an index system for credit risk assessment. After that, the sample data and these indicators are combined using factor and cluster analysis to produce the real credit rating, which is then assigned a category. The decision tree model and the artificial neural network (ANN) model are then compared. The results show that the ANN model is more likely to make a prediction, with an accuracy of 95.807 compared to the decision tree model 82.11.