On the credit evaluation model based on GP and BP
Xu Juan · Journal of Hefei University of Technology · 2010
A two-stage hybrid credit evaluation model is proposed in this paper based on genetic programming(GP) and back propagation (BP) neural network. This model can make up for the shortage of single algorithm and provide a simple rule for the decision maker by using GP to extract IF-THEN rule. It also uses BP neural network to solve the mismatching problem of IF-THEN rule. Simulation experimental results of UCI data set show that the model is feasible and it can improve the accuracy of classification compared with other relative models.