Phishing Email Classification App with Redis

This project demonstrates a phishing email classification application using machine learning. It includes four main components:

• Model Training (train.py): Processes the dataset, trains a model using TF-IDF vectorization and logistic regression, and saves the trained model.
• Model Validation (validate.py): Evaluates the saved model on a test dataset and outputs performance metrics such as accuracy, precision, and recall.
• Model Serving (serve.py): Implements a FastAPI-based service to provide real-time phishing email predictions. Redis caching is used to optimize performance by speeding up repeated requests.
• Redis Check (check_redis.py): A utility script to inspect the Redis database, displaying the number of cached entries and sample predictions.

---

Application Workflow

The following diagram illustrates how the application works, including the interaction between the machine learning model, FastAPI, and Redis caching:

---

Project Structure

.
├── data/
│   └── phishing_emails.csv         # Kaggle dataset 
├── phishing_model.pkl              # Saved model file
├── train.py                        # Script for training the model
├── validate.py                     # Script for validating the model
├── serve.py                        # FastAPI app for model serving with Redis caching
└── check_redis.py                  # Script to inspect Redis cache entries

---

Prerequisites

• Python 3.7+
• Pip
• Redis (see instructions below for running Redis on Windows)

Install the required Python packages with:

pip install -r requirements.txt

---

Dataset

The dataset is sourced from Kaggle. Download the dataset and place it in the data/ directory.

---

Usage

1. Training the Model

Run the training script to process the dataset and generate the model file:

python train.py

2. Validating the Model

Run the validation script to evaluate the model's performance:

python validate.py

3. Serving Predictions

Ensure Redis is running (see instructions below for Windows), then start the FastAPI server:

python serve.py

Access the API at http://localhost:8000 and send POST requests to the /predict endpoint with JSON payloads:

{
  "text": "Your email content here..."
}

4. Checking Redis

Use the check_redis.py script to inspect the Redis cache. This script displays the number of cached entries and the first five email texts along with their predictions:

python check_redis.py

---

Running Redis on Windows

1. Enable WSL: Follow Microsoft's guide to enable WSL and install a Linux distribution (e.g., Ubuntu) from the Microsoft Store.
2. Install Redis: Open your WSL terminal and run:

   sudo apt update
   sudo apt install redis-server

3. Start Redis:

   sudo service redis-server start

4. Access Redis from Windows: With Redis running in WSL (default on localhost:6379), the Python Redis client in your FastAPI app can connect without extra configuration.

---

API Usage

Send a POST request to http://localhost:8000/predict with the following JSON payload:

curl -X 'POST' \
  'http://localhost:8000/predict' \
  -H 'accept: application/json' \
  -H 'Content-Type: application/json' \
  -d '{
  "text": "todays floor meeting you may get a few pointed questions about today article about lays potential severance of $ 80 mm"
}'

The API responds with a JSON object containing:

• prediction: The predicted class name (e.g., "Phishing Email" or "Safe Email").
• probability: The confidence score of the prediction.

Example response:

{
  "prediction": "Safe Email",
  "probability": 0.7791625553383463
}