Non-Linear Time Sequence Problems Based on GA-BP Neural Network Prediction Model

In the fields of image recognition and computer vision, data mining and predictive analytics, BP neural network as a kind of simple structure and powerful model has been widely concerned by scholars, and the improvement of BP neural network is very important. In this paper, we used genetic algorithm to optimize BP neural network, and optimize the selection of weights and thresholds of BP neural network through the characteristics of local optimization of genetic algorithm. Compared with the single BP neural network, the root mean square error (RMSE) of the optimized BP neural network had been reduced by 0.82%, the mean absolute error (MAE) had been reduced by 0.85%, and the coefficient of determination had been improved by 1.34%• The experimental results have shown that the optimization of BP neural network with genetic algorithm is more advantageous than the single BP neural network, and the optimized BP neural network has a strong ability to deal with the complex problems with high dimensions and multi-objectives.