Generative AI based Fertilizer and Pesticide Recommendation System for Farmers

Kalaiselvi, G. and Nisha, M. and Rudhra, M. and Shrivaishnavi, V. (2025) Generative AI based Fertilizer and Pesticide Recommendation System for Farmers. International Journal of Science and Research Archive, 15 (2). pp. 1536-1546. ISSN 2582-8185

Agriculture is the backbone of the Indian economy, with 60% - 70% of the Indian people relying on agriculture for livelihood. Limited education and awareness hinder farmers from understanding the effects of improper fertilizer and pesticide use, leading them to heavily rely on agricultural experts. However, this expert advice often meets fewer of the farmer’s specific needs, such as soil quality, weather conditions, or crop health. Due to the inadequate fulfillment of farmer’s specific requirements from the experts, it results in inefficiencies, excessive use of fertilizers and pesticides, and harmful effects on the environment. With the aid of Generative Artificial Intelligence, the farmers get recommendation on fertilizers and pesticides for the crops and know which crops are best to grow on their land. Generative AI based fertilizer and pesticide recommendation system generate the real-time prediction that analyzes environmental and soil factors like Nitrogen (N), pH, Organic Matter, Microbial activity which recommends the fertilizer by using XGBoost and pesticide recommendation based on the image analysis of the crop affected by pest using Inception V3. When providing inputs on the web interface, the system recommends, what fertilizer to be used and helpful for identification of the pest and prescribe the appropriate dosage of pesticide. Additionally, farmers can contact directly through the GPT-4o-mini service for quick help.