Predictive modeling of crop yield using random forest: A comprehensive analysis of meteorological and agricultural factors

Predicting crop yields is a crucial agricultural undertaking that helps with improved resource management and planning. Because machine learning can scan enormous datasets to find hidden patterns and provide reliable forecasts, it has become a useful tool for addressing the issues given by weather variability in agricultural forecasting. In this work, we assess the effectiveness of several machine learning models, including Multiple Linear Regression, K-Nearest Neighbors (KNN), Random Forest Algorithm, and Support Vector Machine (SVM), for predicting rice crop yield in several districts within the Indian state of Chhattisgarh. Key performance metrics were used to evaluate the models. Among the models, Random Forest Algorithm outperformed the others with an R^2 score of 0.9476, MSE of 0.02254, RMSE of 0.15016, and MAE of 0.0036, indicating a high level of accuracy and precision and low error rates in predicting the crop yield, while simpler models such as Multiple Linear Regression showed comparatively lower predictive power. This analysis highlights the importance of selecting appropriate machine learning algorithms for enhancing the precision of crop yieldforecasts. The complete analysis was done by using open-source software namely R-Software.