📧 Phishing Email Detection API

An AI-powered REST API built with FastAPI that detects whether an email is a phishing attempt or safe using a trained Machine Learning model. This project is containerized using Docker and ready for deployment on platforms like Render or AWS.

---

🚀 Features

• FastAPI-powered REST API with Swagger (OpenAPI) documentation
• Text Preprocessing Pipeline with NLTK for cleaning email content
• ML Model Integration: Random Forest trained on TF-IDF features
• Secure API using Bearer Token authentication
• Detailed Logging with timestamp, IP tracking, and prediction logs
• Dockerized for consistent local and production environments
• Fully Tested using pytest for API endpoint validation

---

🛠️ Tech Stack

• FastAPI for building the REST API
• scikit-learn and joblib for the ML model
• NLTK for text preprocessing
• Docker for containerization
• Render (or AWS) for deployment
• pytest for automated testing

---

🧪 Input Example

Send a POST request to /predict with a JSON body:

{
  "text": "Congratulations! You've won a free iPhone. Click here."
}

If the token is valid, you'll get a prediction:

{
  "prediction": "Phishing Email 🚨"
}

---

📂 Project Structure

phishing_email_detector/
├── app/
│   └── main.py               # FastAPI app
├── utils/
│   ├── text_cleaning.py     # Preprocessing
│   └── logger.py            # Logger setup
├── model/
│   ├── phishing_rf_model.pkl
│   └── tfidf_vectorizer.pkl
├── tests/
│   └── test_main.py         # Pytest-based tests
├── logs/
│   └── app.log              # Auto-generated logs
├── .env                     # Environment variables
├── requirements.txt         # Dependencies
├── Dockerfile               # Docker setup
└── README.md

---

⚙️ Installation

📌 Prerequisites

• Python 3.9+
• Docker
• Git

1. Clone the Repository

git clone https://github.com/yourusername/phishing-email-detector.git
cd phishing-email-detector

2. Set Up Environment

Create a .env file:

API_TOKEN=your-secret-token

3. Install Dependencies

pip install -r requirements.txt

4. Run the App Locally

uvicorn app.main:app --reload

Go to http://127.0.0.1:8000/docs to explore the Swagger UI.

---

🧪 Testing

Run unit tests using:

pytest tests/

---

🐳 Docker Usage

Build the Docker Image

docker build -t phishing-detector .

Run the Container

docker run --name phishing-api -p 8000:8000 --env-file .env phishing-detector

Visit http://localhost:8000/docs to test it in Swagger.

---

☁️ Deploy to Render

1. Push Code to GitHub

Ensure your repo is pushed to GitHub.

2. Create a New Render Web Service

• Go to Render
• Select "New Web Service"
• Connect to your GitHub repo
• Choose environment: Docker
• Set your environment variables (API_TOKEN)

3. Deploy

Render will build and deploy your app. Use the provided Render URL to access Swagger.

---

🔐 API Authentication

Use the Authorize 🔒 button in Swagger to test:

Bearer your-secret-token

You can also test using Postman with the Authorization header:

Key: Authorization
Value: Bearer your-secret-token

---

📝 Logs

Logs are saved under /logs/app.log. Example entry:

2025-04-19 10:45:00 - INFO - Prediction: Phishing Email 🚨 | From: 127.0.0.1

---

🧠 Skills Demonstrated Through This Project

• ✅ Demonstrates End-to-End ML Integration
• ✅ Implements Authentication & Logging
• ✅ Shows MLOps practices (Docker, testing, deployment)
• ✅ Easy to demo with Swagger UI & Postman

---

🤝 Contributing

Feel free to fork and enhance. PRs are welcome!

---

📧 Contact

Deepak B
LinkedIn | deepakbalraj511@gmail.com