Enhancing Agricultural Resilience in India: Leveraging Ensemble Learning for Crop Yield Prediction

Climate change is a major concern globally and India majorly depends upon agricultural crops and its economy. India being an agricultural farmland is majorly affected by the impact environmental changes from last few years. Crop yield and productivity are the significant results which affects the farmers financially when a crop is harvested. Selection of the right crop based upon the predictions is a technological advancement for the farmers, which helps to plan the agricultural products and crops strategically. In this paper, prediction of crops is done by implementing ensemble learning techniques. The advancement of machine learning and advantages of these techniques helps to predict the right crop before harvesting. In this paper, the models used for predicting the crops are Random Forest, AdaBoost, Gradient Boosting, Elastic Net, Decision Tree, K-Nearest Neighbor, Lasso Regression, and Bagging Regression. Highest accuracy was achieved through a combination of Lasso, bagging, KNN, and random forest which is of 95.3%.