Rice and Wheat Yield Prediction in India Using Decision Tree and Random Forest

Abstract

One of the main sources of revenue and growth in Indian economy is from agriculture. It is often a gamble for the farmers to obtain a decent yield, considering the unpredictable environmental conditions. This paper deals with the prediction of the yield of rice and wheat using machine learning algorithms using the annual crop yield production and the annual rainfall in the different districts of India. In this paper, a popular prediction model is developed using algorithms such as decision tree and random forest to predict the yield of most widely grown crops in India like rice and wheat. The features used were the area of production, rainfall, season and state. The season and the state were one hot encoded features. Mean square error was used to measure the loss. The dataset was prepared by combining the crop production in the various states and the rainfall dataset in the respective states. Index Terms : Machine Learning, XGBoost, Decision Tree, Random Forest, Data Preprocessing, Data Visualization, Prediction