Credit scoring model based on neural network

The research develops a neural network credit-scoring model by using the principle of multilayer perceptron. 106 selected listed companies are divided into two groups: "good" ones and "bad" ones. All the data come from China. The simulation results show that the neural network credit-scoring model has a high rate of classification accuracy and strong suitability. We construct the structure of the neural network credit scoring and present its learning algorithm.