PhishGuard is a machine learning-based web application designed to detect and prevent phishing attacks. It analyzes URLs and email-related features to determine whether a link is legitimate or malicious.
- 🔍 Detects phishing URLs using ML models
- 📊 Feature extraction from URLs and email data
- 🌐 Simple frontend interface for testing links
- ⚡ Fast predictions using trained model
- 🧠 Train your own model using provided scripts
PhishGuard/
│
├── backend/
│ ├── app.py
│ ├── extract_email_address_features.py
│ └── train.py
│
├── frontend/
│ ├── index.html
│ └── plotly.min.js
│
├── .gitignore
└── README.md
- User enters a URL
- Backend extracts features
- Machine learning model analyzes patterns
- Returns prediction: Phishing or Safe
git clone https://github.com/AyushI7G/PhishGaurd2.0.git
cd PhishGaurd2.0
pip install -r requirements.txt
python backend/train.py
This will generate the trained model files required for predictions.
python backend/app.py
Open frontend/index.html in your browser.
Dataset is not included due to size and security reasons.
You can use:
- PhishTank dataset
- Kaggle phishing datasets
- Python
- Machine Learning (XGBoost / Scikit-learn)
- HTML, CSS, JavaScript
- Plotly (for visualization)
- Model files are not included—generate them using
train.py - Dataset must be downloaded separately
- Ensure correct file paths while running locally
- Chrome extension integration
- Real-time URL scanning API
- Better UI/UX
- Deployment on cloud
Ayushi Gupta
Feel free to fork this repo and improve it!
This project is open-source and available under the MIT License.