Use of Random Forest Regression Model for Forecasting Food and Commercial Crops of India

¹ University of Alkafeel, Najaf, Iraq
² PSG College of Arts & Science, Coimbatore, India
³ SGT University, Gurugram, India
⁴ Lovely Professional University, Phagwara, India
⁵ JNKVV College of Agriculture, Rewa, India
⁶ South Ural State University, Chelyabinsk, Russia

^* Corresponding Author: ali.j.r@alkafeel.edu.iq

Abstract

Agriculture is the backbone of Indian Economy. Proper forecast of food crops and cash crops are necessary for the government in policy making decisions. The present paper aims to forecast Wheat and Sugarcane yield using Random Forest Regression. For the development of Random Forest models, Yield has been taken as dependent variable and variables like Gross Cropped Area, Maximum Temperature, Minimum Temperature, Rainfall, Nitrogen, Phosphorous Oxide, Potassium Oxide, Minimum Support Price and Area under Irrigation are taken as independent variables for both Wheat and Sugarcane crop. Values of R² for Wheat and Sugarcane is 0.995 and 0.981 which indicates that the model is a good fit and other performance measures are calculated and results are satisfactory.