Crop Management Using Machine Learning

Abstract

The agricultural production plays a vital role in the economy of the country. One of the challenges faced by the farmers is to produce sufficient crop yield to meet the increasing consumer demand while maintaining the crop quality and quantity. Crop yield and quality are greatly influenced by the activities of soil management, water management, disease detection, and weed detection. The applications of Machine Learning (ML) in crop management would result in accurate and appropriate crop yield and quality forecasts. The purpose of this study is to present ML models applied to the crop production and get predicted insights to make the right decisions by the farmers to achieve maximum profit and reduce the risk. We implemented disease detection as case study using ML classifiers and Ensemble Learning methods. We analyzed the performance of these models and observed that Ensemble Random Forest outperformed with the prediction accuracy of 98.12% when compared with the other classifiers.