🌟 This project aims to boost disaster preparedness by predicting natural calamities using Machine Learning 🧠, time-series models like ARIMA 📈, and a blockchain-based relief fund collection gateway 💸 powered by smart contracts on Ethereum 🔗.
-
Natural Calamity Probability Predictor 🌩️:
- Predicts the likelihood of natural disasters, such as cloudbursts, floods, and rainfall 🌧️ for specific dates using advanced ML algorithms like Gradient Boosting 🌲, ARIMA time series analysis 📉, and Haversine Formula 📐.
-
State-Level Precipitation Timeline Videos 🎥:
- Generates GeoTIFF-based precipitation timeline videos across India 🇮🇳, its individual states 🗺️, and even districts in West Bengal 📍. The analysis can provide users with district-wise precipitation patterns, offering granular disaster insights.
-
Blockchain-Based Relief Fund 💵:
- Utilizes Solidity smart contracts to enable secure and transparent Web3 token-based donations. Each transaction is stored on the Ethereum blockchain ensuring complete trust and transparency.
-
District-Wise Disaster Forecasting for West Bengal 📊:
- Users can view district-level forecasting for West Bengal based on historical weather data 🌦️. This fine-grained prediction system analyzes past trends using ARIMA and ML models for highly localized disaster preparedness.
👆 A glimpse of the app interface for predicting the probability of natural disasters using ML and ARIMA-based models.
3734992b607c5b21a8b6b7e0c74b100898707b8cd8a42321648a196a.mp4
🎥 Visualize precipitation timelines across India and its states with GeoTIFF data.
💸 Leverage blockchain-based relief fund collection, ensuring transparent, immutable donations.
📊 Analyze district-level precipitation in West Bengal for a more detailed understanding of weather trends.
To set up HackSynthesis Omicode on your local machine, follow these steps:
-
Clone the repository:
git clone https://github.com/CodenWizFreak/HackSynthesis_Omicode.git cd HackSynthesis_Omicode -
Create a virtual environment (recommended):
python -m venv venv source venv/bin/activate # Windows: `venv\Scripts\activate`
-
Install dependencies:
pip install -r requirements.txt
-
Set up the blockchain environment 🏗️:
- Ensure Node.js is installed.
- Install Truffle and Ganache:
npm install -g truffle npm install -g ganache-cli
-
Compile and deploy smart contracts 📝:
truffle compile truffle migrate --network development
-
Run the Streamlit app:
streamlit run app.py
-
Explore the Features:
- Predict natural disasters: Enter a specific date to receive predictions for cloudbursts, floods, etc.
- Generate Precipitation Videos: Select a region (India, state, or district) and generate a precipitation video based on historical GeoTIFF data.
- District-Level Analysis: Receive a detailed district-wise prediction for West Bengal based on ARIMA and ML models.
- Blockchain Donations: Use the app to make Web3 token transactions toward the relief fund.
-
Blockchain Transactions 💰:
- Make secure Web3 token-based donations for disaster relief using Solidity-based smart contracts.
- Watch live gas fees ⛽ and blockchain confirmations in real-time.
- Comprehensive Disaster Preparedness: Integrates advanced machine learning and time-series models for accurate natural calamity predictions.
- Localized Forecasting: Provides detailed district-level predictions for natural disasters in West Bengal, ensuring targeted insights.
- Blockchain-Based Relief Fund: Utilizes smart contracts on Ethereum for secure, transparent, and traceable donation processes.
- GeoTIFF Visualization: Generates dynamic precipitation timeline videos using GeoTIFF data, enhancing user understanding of weather trends.
- Data-Driven Decision Making: Leverages extensive historical weather data to enhance prediction accuracy and reliability.
- Scalable Framework: The system can be adapted to different regions beyond India, allowing for wider application in disaster management.
- Technological Reliability: Built using proven technologies such as Python, Ethereum, and TensorFlow, ensuring a solid foundation for the project.
- Community Engagement: Encourages active participation through a blockchain-based donation system, fostering trust and local involvement.
- Interdisciplinary Integration: Combines machine learning, geospatial analysis, and blockchain technology for a comprehensive disaster management solution.
- Real-Time Disaster Insights: Offers real-time analytics on disaster probabilities and blockchain transactions, enhancing situational awareness.
- GeoTIFF Data Utilization: Innovatively employs GeoTIFF data to create detailed precipitation timeline visualizations for informed decision-making.
- Localized Impact Focus: Prioritizes district-specific data, highlighting the importance of localized forecasting in effective disaster preparedness.
- Community Empowerment: Equips communities with tools and information necessary for proactive disaster preparedness and response.
- Holistic Disaster Ecosystem: Encompasses the entire disaster management process, from prediction to relief fund allocation, fostering resilience.
- Open-Source Development: Promotes collaborative contributions from developers and researchers to innovate and improve disaster management solutions.
- Awareness and Education: Aims to raise public awareness about disaster preparedness and community involvement through a user-friendly platform.
- Frontend: Streamlit 💻
- Backend: Python (Flask), Streamlit 🚀
- Machine Learning: TensorFlow, Keras 🧠, Scikit-learn, XGBoost, Haversine Formula 📐, ARIMA (AutoRegressive Integrated Moving Average) 📉
- Data Processing: Pandas, NumPy 🧮, GeoTIFF, Rasterio 🌍
- Blockchain: Truffle, Infura, Ganache, Ethereum 🔗, MetaMask 🦊, Web3.py 🌐
- Smart Contracts: ERC-20 Token Standard 💎
- Data Visualization: Matplotlib, Seaborn, Plotly 📊
- Geospatial Data: GeoPandas, Folium 🗺️
- Video Processing: OpenCV 🎥, ImageIO 📅
This project is licensed under the MIT License. See the LICENSE file for details. 📄
Let’s reshape the future of disaster management with advanced machine learning, ARIMA models, geospatial analysis, and blockchain technologies! 🚀