Stacked Regressor for Crop Yield Prediction

A significant number of people in developing nations like India, when a primary source of income is farming, depend on the agriculture sector for their economy. The agricultural sector can be improved to make it more effective and lucrative for farmers by harnessing the prospects of data and analytics. This work suggests using deep learning and machine learning techniques to forecast crop yield throughout India. The work trains the models using a variety of input variables, including weather information, soil characteristics and crop character traits. The findings demonstrate that models based on machine learning and deep learning are effective at accurately predicting crop yield. Intelligent choices about crop production and management can be made using the proposed approach by farmers and policymakers. Using a stacked ensemble model of Random Forest, Gradient Boosting, Bagging regressor and RNN as the meta-learner layer, this work evaluates crop yield.