A Crop Recommendation System Using Ensemble Learning Approach for Sustainable Farming

Abstract

Sustainable farming uses a precise crop recommendation model to optimize yield and resource use. Integrating ensembled machine learning with soil and environmental data, this study applies Principal Component Analysis to identify influential factors among 22 crops grown across the world. An ensembled model combining Random Forest, Naive Bayes and Decision Tree produces 99.24% accuracy, outperforming individual models. A website and a smartphone-based application is built on the proposed model for real-world use by farmers. This data-driven approach supports robust recommendations and enhances agricultural decision-making. This study demonstrates the role of artificial intelligence in advancing resilient and resource-efficient farming, thus providing a practical tool for sustainable crop selection and management.