Incorporating meteorological data and pesticide information to forecast crop yields using machine learning

Vanma, Pavan Kumar and Booma, Joel and Chinnappan, Moses and Macharla, Balakrishna and Kali, Tharun (2025) Incorporating meteorological data and pesticide information to forecast crop yields using machine learning. World Journal of Advanced Engineering Technology and Sciences, 15 (2). pp. 488-495. ISSN 2582-8266

Crop Yield Predictor is a full-stack web application that leverages machine learning to estimate agricultural crop yields based on key environmental and input factors. Using a dataset spanning from 1997 to 2017, the system considers variables such as crop type, season, state, rainfall, temperature, fertilizer usage, pesticide application, and cultivated area. The backend, built with FastAPI, hosts a trained regression model (XGBoost or Random Forest) that predicts crop yield in hectograms per hectare. The frontend, developed using React, allows users to input field data and receive real-time yield predictions along with smart recommendations on pesticide and fertilizer usage. This intelligent advisory system aims to support farmers and agricultural planners in making informed decisions, optimizing resource usage, and enhancing crop productivity through data-driven insights. This system bridges the gap between data analytics and agriculture, promoting smarter resource use and precision farming practices.