Predictive Modelling of Crop Rotation Using Data Mining Approaches

Agriculture is crucial for economic growth and food security, particularly in agro-based countries. As the global population grows, the demand for food increases, necessitating improvements in agricultural productivity. Traditional methods have often fallen short, and innovative approaches such as data mining and machine learning are needed. This research aims to develop a predictive model for crop rotation using machine learning techniques. A comprehensive dataset was collected and preprocessed to train various algorithms. The proposed model demonstrated that machine learning could effectively predict suitable crops for cultivation, thereby enhancing crop yield and sustainability. The evaluation results were promising, with the Random Forest model achieving a precision of 0.67 to 1.00, recall of 0.43 to 1.00, and F1-score of 0.60 to 1.00; the Decision Tree model had a precision of 0.50 to 1.00, recall of 0.43 to 1.00, and F1-score of 0.50 to 1.00; and the K-Neighbors Classifier model showed precision of 0.40 to 1.00, recall of 0.43 to 1.00, and F1-score of 0.50 to 1.00.