Skip to Main Content
The price of agricultural products are affected by many factors, and the relationship between independent variables and dependent variables can not use specific mathematical formula to express. The traditional prediction methods emphasized on the linear relationship between the prices, and the limitation is apparent, which lead to the low prediction precision. This paper proposes an improved BP neural network model. Firstly, get factors of price fluctuation of agricultural products through the qualitative analysis and then use the MIV method to choose the strong influent factors as the input nodes of a neural network. Find the optimal structure of BP network through the improved learning algorithm, and then use the improved model to realize the agricultural high precision simulation of the product price. The results show that, the model provides an effective prediction tool for the agricultural product price forecasting.