Wheat Yield Estimation and Predication Via Machine Learning

Abstract

A precise wheat yield estimation and prediction are significant for food safety and security purposes of a region or a country, which provide societal peace and sustainable development. Earlier methods for wheat yield prediction are time-consuming, site-specific, and expensive, require more manpower, and delay results with numerous errors and uncertainty. This research work uses numerous heterogeneous data in machine learning via linear regression (LR), decision tree (DT), and random forest (RF) regression by python for accurate wheat yield estimation and prediction at 10m resolution. In a comparison of all three regressions, RF shows the highest accuracy with R2: 98, and RMSE: 1.40, which is also increasing from seedling to harvest growth stage. This research work provides precision agriculture for the sustainable development of a region or a country.