Analysis on the Influence of Regional Agricultural Planting Factors on Grain Yield Based on Genetic Feedback Process Neural Network

With the accumulation of Agricultural Big Data, the management and decision-making of agricultural planting production based on single factor expose the problem of lack of comprehensive analysis. In this paper, considering the high dimensions and time-varying characteristics of agricultural big data, a new time series analysis model based on Genetic Feedback Process Neural Network (GFPNN) is proposed to analyze the influence of regional agricultural planting factors on grain yield. Take China as an example, the data of 11 agricultural planting factors in China from 1980 to 2020 constitute a time series. Analyze the time series by GFPNN to establish the influence model of agricultural planting factors on grain yield. Results show that this model can estimate grain yield with an accuracy of 97.83%, which is more accurate and efficient than traditional neural network. It can evaluate the influence of changes in agricultural planting factors on grain yield. It can be a reference for taking measures to deal with the change of grain yield and making grain policy; in the long run, developing and introducing appropriate agricultural machinery, popularizing efficient water-saving irrigation technology, further investing in rural education, and reducing planting costs can improve the grain yield to a certain extent.