In India Agriculture is influenced by the diverse climatic conditions and varied land types, which leads to diversity in cropping patterns and farming methods. In agriculture, cropping patterns depend on several factors like seasons, soil type, climate, and water availability. Due to variations in environmental and agronomic parameters; the selection of best crop for cultivation is a major challenge for farmers. The farmer often faces difficulties in selecting the right crop for sowing, and needs expert’s advice for achieving maximum yield. The integration of agricultural practices with Machine Learning (ML) offers a transformative approach to tackle this challenge. The prediction of best suitable crop plays a significant role in optimizing agricultural output, ensures food security, and manages resources efficiently. This research involves developing hybrid models (HVM1, HSM2, HVM3, HSM4) using diverse classification and regression ML models such as RF, DT, KNN, SVM, LogRes and Stacking and Voting ensemble techniques. The trained models used for predicting best suitable crop to be grown for particular environmental and soil conditions. The potential of Machine Learning made revolution in crop prediction by providing robust, accurate, and actionable forecasts. The HSM4 model developed using diverse classification models and ensemble stacking technique shows the excellent results with 99.4% accuracy. By leveraging HSM4 model, this research help farmers in optimizing resource use, reducing input costs, and improve agricultural productivity and sustainability.
Kadwey et al. (Wed,) studied this question.