Fraud Detection System

A distributed system for real-time transaction fraud detection using microservices architecture.

Project Overview

The Real-Time Fraud Detection System is designed to process transactions in real-time, analyze them for potential fraud, and store relevant data for further analysis. It consists of several key components:

• Transaction Ingestion Service: Handles incoming transaction data and publishes it to Kafka.
• Fraud Analysis Service: Consumes transaction data from Kafka, applies fraud detection algorithms, and interacts with the ML component.
• Python ML Component: Provides fraud scoring based on transaction data.
• Data Storage Integration: Utilizes Couchbase for transactional storage and AWS Neptune for graph-based analysis.
• CI/CD Pipeline: Automates testing and deployment processes using GitHub Actions.

System Architecture

The system consists of three main components:

1. Transaction Service (Node.js/TypeScript)

   • REST API for transaction ingestion
   • Kafka producer for async processing

2. Fraud Analysis Service (Node.js/TypeScript)

   • Kafka consumer for transaction processing
   • Integration with ML component

3. ML Component (Python)

   • Machine learning model for fraud scoring
   • Model training and inference APIs

Architecture Diagram

Project Structure

.
├── transaction-service/     # Transaction ingestion REST API
├── fraud-analysis-service/ # Kafka consumer service
├── ml-component/          # ML model and scoring
├── terraform/            # Infrastructure as Code
└── ci-cd/               # CI/CD configuration

Setup and Installation

Prerequisites

• Node.js 18+
• Python 3.9+
• Docker
• Terraform

Environment Setup

1. Transaction Ingestion Service:

   • Navigate to transaction-service and run npm install.
   • Set up environment variables in a .env file.

2. Fraud Analysis Service:

   • Navigate to fraud-analysis-service and run npm install.
   • Set up environment variables in a .env file.

3. ML Component:

   • Navigate to ml-component and run pip install -r requirements.txt.
   • Set up environment variables in a .env file.

Running the Application

• Transaction Ingestion Service: Run npm start in the transaction-service directory.
• Fraud Analysis Service: Run npm start in the fraud-analysis-service directory.
• ML Component: Run python app.py in the ml-component directory.

CI/CD Pipeline Overview

The CI/CD pipeline is configured using GitHub Actions and includes:

• Linting and testing for Node.js/TypeScript and Python components.
• Building and deployment processes.

Troubleshooting

• Ensure all environment variables are correctly set.
• Check service logs for error messages.
• Verify network connectivity between services.

Infrastructure

Infrastructure is managed using Terraform and deployed on cloud infrastructure.

CI/CD

Continuous Integration and Deployment is handled through GitHub Actions with separate pipelines for each service.