Denguard is an AI-powered Django web application that predicts dengue outbreak risks and recommends smart municipal actions using machine learning. It leverages real-time weather data, historical case reports, and medical inputs to help both citizens and doctors stay alert and respond proactively.
- 📍 Location-Based Risk Prediction using SVM
- 🩺 Doctor Dashboard to log real dengue cases and issue medical advice
- 🌧️ Live Weather Data Integration via OpenWeatherMap API
- 🧠 Action Recommendation System using Markov Decision Process (MDP)
- 🗺️ Heatmap Visualization of dengue outbreak zones by area
- 🔐 Two User Roles: Normal User & Doctor
- 🖥️ Responsive Web Interface with clean, informative design
- Backend: Django, Scikit-learn, Pandas, NumPy
- Machine Learning:
- Support Vector Machine (SVM) for outbreak classification
- Markov Decision Process (MDP) for decision policy
- Frontend: HTML, CSS, JavaScript, Bootstrap/Tailwind
- Data Handling: Real-life Dengue Cases + Weather Forecast APIs
- Visualization: Heatmaps, tables, and dynamic recommendations
- DGHS Bangladesh — Real dengue case reports
- OpenWeatherMap API — Forecasted rainfall, temperature, humidity
- Remote Sensing / Public Reports — Waterlogging and flood-prone areas
- Manual Doctor Inputs — Hospital-reported confirmed dengue cases
Contributions are welcome! Feel free to:
- Submit a pull request
- Open an issue
- Suggest improvements or features
Whether you're a public health expert, developer, or designer—your help is appreciated!
This project is licensed under the MIT License. See the LICENSE file for details.
Author: Maruf Hossain
Department: Computer Science & Engineering
University: Green University of Bangladesh
📧 Email: maruf.bshs@gmail.com, 221902318@student.green.edu.bd