Research of Population Prediction Based on GA-BP Neural Network

The BP neural network is a feed-forward network trained by backward propagation of errors algorithm, which is the most widely used neural network model, but BP neural network can´t avoid the shortcoming that is strong randomness and easily converging to local minimum. In this paper, we discuss the principle, structure and realization way of BP neural network improved by genetic algorithm, which can effectively improve the performance of BP neural network. At the same time, the improved model is used in the tax case-selecting, the financial statements and tax returns of 80 enterprises are analyzed, and then the analysis result is compared with that of BP neural network and binary logistic regression analysis. The comparing analysis shows that the GA-BP neural network method can assist the case-selecting and can improve the efficiency and effect of the tax inspection.