Application of BP neural network optimization algorithm based on genetic algorithm in credit risk early-warning of commercial bank
Jie Su, Yaning Zhang · 2017
In order to study the early warning system of commercial bank's credit risk, a new early-warning model is established by optimizing the weights and thresholds of BP neural network using genetic algorithm based on nonlinear combinatorial forecasting principle. Finally, the training set and test set errors of BP neural network under stochastic weights and thresholds and optimal weights and thresholds have been compared. The BP neural network optimization algorithm based on genetic algorithm can predict the credit risk of commercial banks more effectively.