Crop Yield Prediction Using ML

Abstract

India’s agriculture sector is pivotal to the nation’s economy and sustains livelihoods for millions. With diverse agro- climatic zones, India boasts a rich agricultural heritage encom- passing crops like rice, wheat, sugarcane, and cotton.For farmers, decision-makers, and other stakeholders to allocate resources and ensure food security, accurate crop yield prediction is essential. This study looks into how machine learning algorithms might be used to increase the precision of crop yield forecasts in India.The study looks at how machine learning models can take into account a number of variables that impact crop yields, such as crop type, season, state, area, fertilizer, pesticide, and rainfall. The effectiveness of various algorithms, such as LinearRegression, Lasso, Ridge and DecisionTreeRegressor, is evaluated.Out of the three Machine Learning methods, the DecisionTreeRegressor algorithm demonstrated the best performance, as seen by its lowest MAE (mean absolute error) value and highest R² value. These findings imply that machine learning algorithms have the potential to greatly increase agricultural yield projections’ accuracy in Morocco, which might enhance food security and maximize farmers’ use of available resources.