🛡️ Phishing Website Detection System using Machine Learning and MLOps

This project involves developing a machine learning system to predict whether a website is phishing or not. The system is designed to enhance cybersecurity by identifying and blocking malicious websites. The project integrates machine learning techniques with MLOps tools to create a scalable, automated, and efficient solution.

🚀 Project Overview

• 🎯 Objective: Predict phishing websites using machine learning to improve cybersecurity.
• 🛠️ Key Technologies: Python, Docker, AWS, MongoDB, FastAPI, GitHub Actions, MLflow, Airflow, Terraform.

🌟 Features

• 🔍 Machine Learning Model: A model trained to detect phishing websites based on various indicators.
• 🔄 MLOps Pipeline: Automated deployment pipeline using GitHub Actions, MLflow, Airflow, and Terraform.
• ⚙️ Scalable Deployment: The system is deployed on AWS using Docker, and MongoDB, ensuring scalability and efficient management.
• 🌐 API Integration: FastAPI is used to build and deploy APIs for interacting with the model.
• 💾 Database Integration: MongoDB is used for storing and managing website data during model training and deployment.

🛠️ Installation and Setup

Prerequisites

• 🐍 Python 3.8+
• 🐳 Docker
• ☁️ AWS account with necessary permissions
• 🍃 MongoDB

🖥️ Setup Virtual Environment

python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txt

⚙️ Configure AWS

• Ensure your AWS CLI is configured with the correct credentials.
• Update the terraform/variables.tf file with your AWS settings.

🚀 Deploy the System

1. 🐳 Build Docker Image:

   docker build -t phishing-detection:v1 .

2. 📤 Push to AWS ECR: Follow AWS ECR instructions to push your Docker image.
3. 🛠️ Run Terraform:

   cd terraform
   terraform init
   terraform apply

🧑‍💻 Usage

1. 📊 Training the Model:

   • Prepare your dataset and upload it to the specified MongoDB collection.

2. 🔍 Prediction:

   • After deployment, the system will automatically predict whether new websites are phishing and store the results in MongoDB.
   • Interact with the model via APIs built with FastAPI.

📈 Results and Evaluation

• The model's performance metrics, such as accuracy, precision, and recall, are tracked using MLflow.

🤝 Contributing

If you'd like to contribute to this project, please fork the repository and use a feature branch. Pull requests are welcome.

📧 Contact

For any questions or feedback, please reach out to Krishna Shukla.