Analysis on Crop Yield Prediction using various Ensemble Methods

Agriculture is a crucial facet of human existence as it fulfils the primary and fundamental requirement of sustenance, namely food. Therefore, it is incumbent upon us, as human beings, to cultivate and enhance our agricultural methodologies. Crop Yield Prediction is a method that enables farmers to determine the most suitable crops to produce in specific seasons based on historical data, prior to harvesting. This research examines the prediction of crop productivity in three distinct states of India. The states are Bihar, Jharkhand, and Odisha. We employ many ensemble methods, namely Random Forest, XGB Regression, and Stacking Regression, in which we combine Random Forest and XGB Regression. Next, compute the results by evaluating L1 and L2 using LASSO and Ridge Regression. Ultimately, XGB Regression yields a more precise outcome in comparison to the other two methods.