AI-Powered Agriculture Chatbot for Smart Farming Assistance

Abstract

Agriculture plays a vital role in the economic development of many countries, especially in regions where farming is the primary source of livelihood. However, farmers often face challenges such as unpredictable weather conditions, lack of awareness about suitable crops, pest infestations, and limited access to expert advice. This paper presents an intelligent agriculture chatbot integrated with a crop prediction system using Convolutional Neural Networks (CNN), Random Forest, and Support Vector Machine (SVM) classifiers. The chatbot component employs Natural Language Processing (NLP) with TF-IDF vectorisation and cosine similarity to provide real-time conversational assistance on crop selection, fertiliser usage, pest control, irrigation, and market guidance. The CNN-based crop prediction module analyses environmental and soil parameters—including soil type, climate, season, and geographic location—to recommend the most suitable crop for cultivation. Comparative evaluation of three models reveals that CNN achieves 95% accuracy, Random Forest achieves 90%, and SVM achieves 80% on the crop prediction task. The hybrid system—combining deep learning prediction with NLP-driven chatbot interaction—provides a comprehensive, accessible, and scalable smart farming platform that bridges the technology gap for rural farmers.