ML Based Seed And Soil Quality Analysis With Environmental Impacts With Adaptive Recommendations With Chatbot Assistance

This work presents a Hybrid Artificial Intelligence–based agricultural decision support system designed to analyze seed quality, soil health, and environmental conditions in an integrated manner. The system automates the agricultural diagnostic pipeline by combining image-based analysis using deep learning models with real-time environmental data retrieved from weather APIs. A dedicated triage layer fuses machine learning outputs and environmental signals to assess severity and context, enabling adaptive and location-aware recommendations. A large language model–based reasoning module generates structured diagnostic reports and actionable farming guidance, while a chatbot assistant allows users to retrieve historical analyses and interact with the system through natural language. The proposed microservices-based architecture ensures scalability, modularity, and secure data management, supporting precision agriculture and sustainable decision-making.