AI-Powered Phishing URL Detection System using Machine Learning
PhishGuard is an intelligent phishing URL detection system that analyzes website URLs and predicts whether they are Safe or Phishing using Machine Learning. The application extracts meaningful URL-based features, evaluates potential security risks, and provides an intuitive dashboard for users.
- 🔍 Detects phishing URLs in real time
- 🤖 Machine Learning based prediction
- 📊 Displays phishing probability
- 📈 Interactive dashboard
- 📋 Recent scan history
- 🌐 Supports single and multiple URL analysis
- 🛡️ Highlights suspicious URL characteristics
- 📉 Displays extracted URL features
- ⚡ Fast and lightweight Streamlit application
- Python
- Streamlit
- Scikit-learn
- Pandas
- NumPy
- Imbalanced-learn
PhishGuard/
│
├── app.py
├── train_model.py
├── model.pkl
├── dataset.csv
├── requirements.txt
├── README.md
├── screenshots/
└── venv (ignored)
Clone the repository
git clone https://github.com/Harini-Nagireddy/PhishGuard.gitMove into the project
cd PhishGuardCreate a virtual environment
python -m venv venvActivate it
venv\Scripts\activateInstall dependencies
pip install -r requirements.txtstreamlit run app.pyThe application opens automatically at
http://localhost:8501
- Load phishing URL dataset
- Extract URL features
- Train Random Forest classifier
- Save trained model
- Predict URL legitimacy
- Display prediction and phishing probability
- URL Length
- Domain Length
- Number of Subdomains
- HTTPS Usage
- Presence of IP Address
- Special Characters
- Suspicious Keywords
- Query Parameters
- Shortened URLs
- Browser Extension
- Real-time URL reputation lookup
- VirusTotal API integration
- Email phishing detection
- QR code scanning
- Deep Learning model
- Live threat intelligence
Harini Nagireddy
Final Year B.Tech (CSE - Data Science)
Give it a ⭐ on GitHub!