radheyyyyy/HackRust-1.0-2026

# Real Time Phishing Detection System ## Hackathon Details * **Event:** HackRust 1.0 * **Date:** 28–29 March 2026 * **Location:** DCRUST, Sonipat, Haryana * **Team Name:** Hakuna Matata ## Team Members * **Rajyavardhan Radhey** 3rd Year, CSE-AI CSJMU, Kanpur * **Bhumika** 2nd Year, MEE CSJMU, Kanpur ## Project Overview This project is a **real-time phishing detection system** that combines **Machine Learning, rule-based analysis, and domain intelligence** to detect malicious URLs. It consists of: * **Chrome Extension** → Real-time protection while browsing * **Backend (Node.js)** → Decision engine * **ML Model (Python)** → Pattern detection * **React Frontend** → URL analysis dashboard ## How It Works Our system uses a **hybrid detection approach**: ### 1. Rule-Based Engine (45–55%) Detects: * Suspicious keywords (login, verify, secure) * Unicode & homograph attacks * Punycode (`xn--`) * IP-based URLs * Suspicious TLDs (.tk, .xyz, .zip, etc.) * Brand impersonation ### 2. Machine Learning Model (25–30%) Uses **35+ features**, including: * URL length & structure * Special character count * Digit ratio * Entropy (randomness detection) * Domain patterns ### 3. Domain Intelligence * **WHOIS Lookup** → Domain age detection * **Traffic Rank** → Popular vs unknown sites * **Tranco List** → Trusted top domains ### Final Decision The system calculates a **final risk score** and classifies URLs as: * 🟢 Safe * 🟡 Suspicious * 🔴 Phishing ## Features * ✅ Real-time phishing detection * ✅ Chrome extension with live scanning * ✅ Automatic blocking of phishing sites * ✅ Explainable AI (shows reasons) * ✅ Confidence score display * ✅ Unicode & homograph attack detection * ✅ Hybrid ML + Rule-based system ## Tech Stack * **Frontend:** React (Vite) * **Backend:** Node.js, Express * **ML Model:** Python, Scikit-learn * **Extension:** Chrome Extension APIs ## Installation & Usage ### 1. Clone Repository git clone cd HackRust-1.0 ### 2. Backend Setup cd Backend npm install node server.js Server runs on: http://localhost:5000 ### 3. ML Model Setup Make sure Python dependencies are installed: pip install joblib scikit-learn pandas Ensure: * `model.pkl` * `scaler.pkl` are correctly placed. ### 4. React Frontend cd phishing-frontend npm install npm run dev Open: http://localhost:5173 ### 5. Chrome Extension 1. Go to: chrome://extensions/ 2. Enable **Developer Mode** 3. Click **Load Unpacked** 4. Select extension folder ## How to Use ### Manual Testing (Frontend) * Enter any URL * Click **Scan** * View: * Result (Safe / Suspicious / Phishing) * Score * Reasons ### Real-Time Protection (Extension) * Open any website * Extension automatically scans URL * Shows popup: * 🟢 Safe * 🟡 Suspicious * 🔴 Blocks phishing sites ## Limitations * WHOIS lookup may fail for some domains * Traffic rank is currently simulated * Requires backend to be running locally ## Future Improvements * Real traffic rank API integration * Advanced deep learning model * Cloud deployment * Browser support beyond Chrome * Threat intelligence integration ## Conclusion This project demonstrates a **scalable, real-time phishing detection system** combining **AI + cybersecurity principles** to enhance user safety while browsing. ## Acknowledgment Built during **HackRust 1.0** with the aim of creating a practical and deployable cybersecurity solution.

标签：自定义脚本