AI-Driven Agricultural Advisory Platform for Odisha

A Django-based web application that provides AI-powered crop yield predictions and personalized agricultural recommendations for farmers in Odisha.

Features

• Yield Prediction: ML-based crop yield forecasting with confidence intervals
• Personalized Recommendations: Top 3 priority actions for irrigation, fertilization, and pest management
• District-Specific: Covers all 30 districts of Odisha
• Multi-Crop Support: Rice, Maize, Wheat, Groundnut, Mung, Cotton, Sugarcane, Turmeric
• Seasonal Advisory: Kharif, Rabi, and Zaid season-specific guidance
• Responsive Design: Bootstrap-based UI for mobile and desktop

Quick Start

1. Install Dependencies:

   pip install Django==4.2.7

2. Run Migrations:

   python manage.py migrate

3. Start Server:

   python manage.py runserver

4. Access Application: Open http://localhost:8000 in your browser

Project Structure

agri_platform/
├── advisory/                 # Main Django app
│   ├── models.py            # Database models
│   ├── views.py             # View controllers
│   ├── forms.py             # Django forms
│   ├── ml_model.py          # ML prediction logic
│   └── urls.py              # URL routing
├── templates/advisory/       # HTML templates
│   ├── base.html            # Base template
│   ├── home.html            # Landing page
│   ├── farm_input.html      # Input form
│   ├── recommendation.html  # Results page
│   └── about.html           # About page
├── agri_platform/           # Django project settings
├── combined_tables.xlsx     # Training data (to be added)
└── manage.py               # Django management script

Usage

1. Home Page: Overview of platform features
2. Farm Input: Enter farm details (district, crop, season, etc.)
3. Get Recommendations: AI generates yield predictions and actionable advice
4. Download/Share: Export or share recommendation reports

Input Parameters

• Location: District selection (30 districts of Odisha)
• Crop: Rice, Maize, Wheat, Groundnut, Mung, Cotton, Sugarcane, Turmeric
• Season: Kharif, Rabi, Zaid
• Field Details: Area, irrigation type, soil type
• Farming Practices: Seed variety, soil health card availability
• Current Status: Pest/disease presence

ML Model Features

The prediction model considers:

• Historical yield data
• Rainfall patterns and weather
• Irrigation infrastructure
• Soil characteristics
• Fertilizer usage patterns
• Seed variety and quality
• Seasonal factors
• District-specific conditions

Technology Stack

• Backend: Django 4.2.7, Python
• Frontend: HTML5, CSS3, Bootstrap 5
• Database: SQLite (development)
• ML: Simplified rule-based system (expandable to scikit-learn)

Future Enhancements

• Integration with real weather APIs
• Advanced ML models (Random Forest, XGBoost)
• Satellite imagery analysis
• Multi-language support (Odia)
• SMS/WhatsApp integration
• Mobile app development

Data Integration

To use with actual data:

1. Place combined_tables.xlsx in the project root
2. Update ml_model.py to load and process the Excel data
3. Train ML models with historical yield data
4. Integrate weather APIs for real-time data

Contributing

1. Fork the repository
2. Create feature branch
3. Make changes
4. Test thoroughly
5. Submit pull request

License

This project is developed for agricultural development in Odisha.