The research on sales forecasting based on rapid BP neural network

The BP neural network and the conjugate gradient algorithm to improve BP neural network algorithm were discussed in this paper. And the simulation results of two kinds of sales forecasting were given. The results indicated that the BP neural network training time was longer and the forecasting precision was lower, but the improved BP neural network algorithm with Conjugate gradient algorithm shorten the training time, and it enhances the forecasting precision. The calculation results showed that the approaches of the improved BP neural network with conjugate gradient algorithm was feasible and could be used for sales forecasting.