AI-Based System for Smart Crop, Fertilizer and Pesticide Recommendation

Agriculture forms the foundation of food security worldwide, but challenges such as overuse of chemical fertilizers, improper pesticide application, and declining soil health hinder sustainable farming. This paper presents an AI-based system for smart fertilizer and pesticide recommendation, designed to assist farmers in adopting precision farming practices. The system collects data from soil tests, crop requirements, and weather patterns, and applies machine learning models to predict the optimal type and dosage of fertilizers and pesticides. Unlike conventional advisory methods that provide generic suggestions, this intelligent system adapts recommendations to local conditions. A decision-support platform has been developed using machine learning techniques such as Random Forest, Decision Trees, and CNN-based pest classification. Experimental results on real-world agricultural datasets show that the system achieves an accuracy of 90% for fertilizer recommendations and 87% for pesticide recommendations, thereby reducing costs and improving crop yield. By bridging the gap between AI research and practical farming, the project contributes to sustainable agriculture, resource optimization, and farmer empowerment.