A Flask-based campus safety management system that provides crime data analysis, danger scoring, and predictive forecasting for university campuses.

• Crime Data Management: Process and serve campus crime incident data
• Danger Scoring: Calculate location-based danger scores and risk assessments
• Predictive Analytics: Machine learning-powered crime forecasting using Facebook Prophet
• Hot Zone Detection: Identify high-risk areas based on historical crime patterns
• Safety Reports: Generate comprehensive PDF safety reports
• AI-Powered Insights: Gemini AI integration for intelligent crime pattern analysis

• GET /data.json - Retrieve crime data
• GET /incidents/all - Get all incident records
• GET /incidents/update - Update database with latest incidents
• GET /api/locations - Get list of all campus locations
• GET /api/clean-data - Clean and normalize crime data

• POST /api/danger-score - Get danger score for specific location
• POST /api/danger-score-by-hour - Get hourly danger scores
• POST /api/danger-forecast - Generate danger forecasts for locations
• POST /api/risk-predict - Predict risk level for location and time
• GET /api/hot-zones - Identify high-crime areas

• POST /api/forecast-prophet - Facebook Prophet-based incident forecasting
• GET /api/crime-forecast - Comprehensive crime forecast with coordinates
• POST /api/compare-days - Compare safety between different days/periods

• POST /api/gemini-summary - AI-generated incident summaries
• POST /api/test-gemini - Test Gemini AI integration
• POST /api/export-pdf - Generate PDF safety reports

• Backend Framework: Flask 2.3.2
• Database: PostgreSQL with psycopg2
• AI Integration: Google Gemini AI
• Machine Learning: Facebook Prophet, scikit-learn
• Data Processing: pandas, numpy
• Geospatial: geopandas, geopy
• PDF Generation: ReportLab
• Environment Management: python-dotenv

1. Install dependencies:

pip install -r requirements.txt

2. Set up environment variables in .env:

GEMINI_API_KEY=your_gemini_api_key
DB_USER=your_db_user
DB_PASSWORD=your_db_password
DB_HOST=your_db_host
DB_PORT=your_db_port
DB_NAME=your_db_name

3. Run the application:

python app.py

backend/
├── app.py              # Main Flask application
├── config.py           # Configuration management
├── prompts.py          # AI prompt templates
├── gemini.py           # Gemini AI integration
├── danger_score.py     # Danger scoring algorithms
├── danger_forecast.py  # Danger forecasting logic
├── hot_zones.py        # Hot zone detection
├── database/
│   ├── db.py          # Database connection
│   ├── incidents.py   # Incident management
│   └── update_database.py # Database updates
├── ml/
│   ├── forecast_prophet.py # Prophet forecasting
│   ├── risk_classifier.py # Risk classification
│   └── plot_danger_by_hour.py # Visualization
├── static/            # Static assets (images, JS)
├── templates/         # HTML templates
└── tests/            # Test suite

curl -X POST http://localhost:5000/api/danger-score \
  -H "Content-Type: application/json" \
  -d '{"location": "Library"}'

curl -X POST http://localhost:5000/api/forecast-prophet \
  -H "Content-Type: application/json" \
  -d '{"location": "Student Union", "days": 7}'

curl -X POST http://localhost:5000/api/export-pdf \
  -H "Content-Type: application/json" \
  -d '{"summary": "Monthly safety report content..."}'

• The application runs in debug mode by default
• CORS is enabled for cross-origin requests
• Static files are served from /static/ directory
• Templates use Jinja2 templating engine

Run tests with:

python -m pytest tests/

1. Follow existing code style and patterns
2. Add tests for new functionality
3. Update documentation as needed
4. Ensure all endpoints handle errors gracefully