The machine learning method for crop and fertilizer recommendations is presented in this study in an effort to improve agricultural productivity. By combining the strengths of XGBoost and Random Forest algorithms through an ensemble model, this system provides customized recommendations based on important soil and environmental parameters, such as temperature, humidity, rainfall, pH, nitrogen, phosphorus, potassium, and other indicators for the optimal crop growth and yield. Using a soft voting technique, the model outperforms traditional single-model systems by averaging predictions from both algorithms to attain a high level of accuracy beyond 98%. This ensemble design captures the high accuracy of XGBoost with its data-handling capabilities while balancing it with Random Forest’s strength and reduced overfitting and also making the model in many different kinds of agricultural environments.
Precision and computational efficiency were demonstrated by the model, which was trained on a dataset of 3,610 samples with an additional 903 samples set aside for testing. Although it takes longer to train, XGBoost helps by producing extremely accurate predictions. Random Forest helps by training more quickly and maintaining performance even when dealing with many different kinds of data. When combined, these models produce an efficient system that can be adjusted to various soil types and climates, empowering farmers to make data-driven, well-informed decisions for sustainable farming.
This study demonstrates how ensemble learning can revolutionize precision agriculture by assisting farmers in making the best decisions possible to enhance crop yield and resource management while encouraging sustainable practices. Real-time data integration may be incorporated into future research to increase the model's adaptability to shifting agricultural conditions. All things considered, this system improves the use of machine learning in the agricultural sector by offering farmers a strong tool to help them in making profitable and sustainable agricultural decisions.