AGRIBOT: AN INTELLIGENT CHATBOT FOR FARMERS WITH CROP RECOMMENDATION AND DISEASE PREDICTION USING MACHINE LEARNING

Abstract: Agriculture remains one of the most important sectors contributing to economic growth and food security across the world. In developing countries such as India, a large percentage of the population depends directly or indirectly on agriculture for their livelihood. Despite technological advancements, farmers continue to face challenges such as improper crop selection, lack of agricultural expertise, unpredictable environmental conditions, and delayed disease identification. These issues often result in reduced productivity, crop loss, and financial instability. To address these challenges, this paper proposes Agribot, an intelligent agricultural chatbot that integrates Machine Learning and Deep Learning techniques to provide personalized crop recommendations and plant disease prediction. The system utilizes a Random Forest Classifier to recommend suitable crops based on soil nutrient parameters and a Convolutional Neural Network (CNN) to identify plant diseases from leaf images. A chatbot interface allows farmers to interact with the system in a simple and user-friendly manner. Experimental results demonstrate that the crop recommendation model achieves an accuracy of 92.7%, while the disease prediction model achieves an accuracy of 94.2%. The proposed system supports smart farming practices by providing timely, accurate, and accessible agricultural guidance.

Keywords: Agriculture, Machine Learning, Deep Learning, Crop Recommendation, Disease Prediction, Random Forest, CNN, Chatbot.