Early Prediction of Crop Yield Using Machine Learning Techniques

Agriculture is the backbone of the world economy, across the business, even if the pay is less for the farmers. Farmers face difficulties yielding the crop in their fields due to new pesticides and new crop varieties. To overcome this problem, develop a machine-learning technique to enhance their financial security by determining which crop generates the highest return. The dataset was taken from the Food and Agriculture Organization and the World Data Bank and compared with different machine-learning techniques, including random forest, KNN, and decision tree regressor, to generate a reliable crop yield prediction. Standard practice dictates that 80% of the data be utilized for model training and 20% for testing. The Random Forest method achieves 97% accuracy, the decision tree regressor achieves 95.7% accuracy, and, in contrast, the KNN method achieves only 20% accuracy for the early prediction of crop yield. Random forest is more accurate than both KNN and decision tree regression.