9th International Conference on ICT for Sustainable Development

Authors - Sandip Shinde, Vaishnavi Gosavi, Revati Nimbalkar, Vaishnavi Patade, Kaustubh Lanke, Pranav Palekar
Abstract - For Maharashtra, India, to optimize agricultural practices and ensure food security, accurate crop yield forecasts are essential. In light of easily accessible metrics including area sown, temperature, rainfall data, and district, this study examines the effectiveness of machine learning algorithms in predicting crop output. Comparison of three well-known algorithms—linear regression, random forest, and decision tree—is the subject of this study. Random Forest proved to be the most effective, achieving the highest accuracy of 98.15%. The study tries to determine the optimal approach for yield prediction by assessing these algorithms' performance on historical agriculture data of Maharashtra such as crop yield, area, production, rain, temperature, etc. In the end, this will boost agricultural output and resistance to climate unpredictability by providing farmers with insightful information for managing resources and choosing crops.