An approach for crop yield prediction using hybrid XGBoost, SVM and C4.5 classifier algorithms

Pecho R.D.C.; Vijaya K.S.; Sharma N.; Pal H.; Jose B.K. · Engineering and Applied Science Research (2024) · DOI: 10.14456/easr.2024.29

Resumen

Science and technological knowledge advancements have resulted in a vast number of data for the agricultural industry. Crop yield prediction (CYP) is a problematic issue in the agricultural field. The proposed work constructs a hybridization of the XGBoost-SVM-C4.5 framework to forecast crop yield. Also, the proposed methodology is employed to predict different crop yields based on temperature, rainfall, and soil parameters. The experimental setup was based on data gathered from the Indian Meteorological Department for various crops. Five metrics were analyzed to determine the performance of each approach under research: Determination Coefficient, Root Mean Squared Error, Correlation coefficient, Mean Absolute Error and Mean Squared Error. Experiments have been conducted to determine the most effective approach for predicting various crop yields. Additionally, the results of this research give an appropriate method for forecasting the future of food production, allowing for a more precise plan for agricultural production.