- Train LSTM based forecasting model on a crop yield dataset.
- Predict the crop yield for the next year for the selected crop type in the selected location
- Pass the yield history and prediction to LLM via LangChain to get actionable insights
- Develop frontend using ReactJS
- Integrate Google Maps API at the frontend
- Develop a chat interface
- Create Flask APIs to bridge frontend and backend functionalities.
- Farmers can rely on precise yield forecasts, minimizing guesswork.
- Helps in understanding and preparing for bad yield years, protecting against severe financial fallout.
- Efficient use of resources (seeds, water, fertilizers) based on targeted insights, reducing waste and expenditure.
- Possibilities of scaling to new regions, crops, or functionalities.
- Update LLM’s data for richer and more precise soil insights.
- Improvements in data, model, or user interaction.