Intelligent Computational Techniques for Crop Yield Based On BP Neural Network

Predicting crop yield is one of the most difficult problems in precision agriculture and one of the largest areas of agriculture to improve the economy. Although, it is a new concept in agriculture. However, there are several challenges due to its dependence on multiple factors such as extreme weather, crop genotype, environmental factors, obtaining minimized production due to climate change, availability of water, management practices, and their environment. Therefore, we designed an efficient BP neural network to obtain crop yield prediction based on environmental data. Hence in this research, we focus on using archival data to predict corn yield across the US cornfield (including 39 states) from 1980 to 2018. The BP neural network model performed very well and achieved a mean-square-error (MSE), mean-absolute-error (MAE), and mean-absolute-percentage-error (MAPE) with an overall accuracy (OA) of 89% of their respective average yields, which is much better than the tested RNNs model, which shows that our BP neural network model has high accuracy in agricultural data prediction.