💳 Online Banking Application – FinTech Software

An end-to-end Online Banking System that mimics real-world banking platforms like ICICI, BOI, or HDFC. This system allows users to securely sign up, authenticate, send/receive money, and visualize transaction history. It also includes an advanced Fraud Detection System to identify suspicious transactions based on historical patterns.

---

📚 Table of Contents

• 🔧 System Architecture
• 🚀 Features
• 📂 Project Structure
• 🔐 Authentication & Security
• 🐳 Docker Setup
• 🛠️ Getting Started
• 📦 Tech Stack
• 🖼️ Application Screenshots
• 📊 Fraud Detection System
• 🤝 Contributors
• 📎 Useful Links

---

🔧 System Architecture

Before development, we invested several weeks designing a robust architecture covering:

• ✅ Functional & Non-Functional Requirements
• 🧬 High-Level Design (HLD)
• 🛠️ Low-Level Design (LLD)
• 🔄 Sequence Diagrams & Flow Diagrams

📌 Design Docs & Diagrams

---

🚀 Features

✅ Backend (Spring Boot + RabbitMQ)

• JWT-based secure login/signup
• Asynchronous transaction processing via RabbitMQ
• Endpoints for user data, transactions, and token validation
• API weight limiting and pagination support

🎨 Frontend (Vite + React + TypeScript)

• Fully responsive & modern UI using TailwindCSS and ShadCN
• Signup/Login, Dashboard, Transaction History
• Money Transfer (Send/Receive) with real-time feedback
• Reusable components and modular code structure

🔍 Fraud Detection

• Machine Learning model that predicts fraud based on transaction patterns
• REST API for integrating predictions into transaction flow
• Built with Python + Flask and explained with Google Gemini

---

📂 Project Structure

FinTech-Software/
│
├── java_backend/
├── Online-Banking-Application-Frontend/
├── ML-Model-For-Fraud-Detection-and-Risk-Analysis-API/

---

🔐 Authentication & Security

• Implemented full token-based security (JWT + Refresh Tokens)
• Secure password storage using hashing
• API protection via route-level guards
• Session timeout and token expiry handling

---

🐳 Docker Setup

To ensure a consistent and collaborative development environment for all team members, we used Docker Compose to bundle our backend environment into a single unit.

🧱 Docker Images Used:

• java_backend-backend:latest — Spring Boot backend
• mysql:8 — Database container
• rabbitmq:3-management — Queue for async transactions

🧩 Docker Compose Container:

All the above services are composed into one container named: java_backend

▶️ Run Docker Compose

docker compose up -d

• This command will build the services and run them on the specified ports if not already available.
• Ensures backend + MySQL + RabbitMQ services are networked and running.

📌 Prerequisite: Docker must be installed and running on your system.

---

🛠️ Getting Started

Here’s how to set up each part of the application.

✅ Backend Setup (Spring Boot + Docker)

📌 Prerequisites: Docker must be installed

1. Clone the backend repo:

   git clone https://github.com/FinTech-Software/java_backend.git
   cd java_backend

2. Run Docker Compose:

   docker compose up -d

This command builds the java_backend container and spins up all services (Spring Boot, MySQL, RabbitMQ) in one go.

---

✅ Frontend Setup (React + Vite + TypeScript)

📌 Prerequisite: Node.js (v18+) installed

1. Clone the frontend repo:

   git clone https://github.com/FinTech-Software/Online-Banking-Application-Frontend.git
   cd Online-Banking-Application-Frontend

2. Install dependencies:

   npm install

3. Start the development server:

   npm run dev

---

✅ ML Model Setup

The fraud detection model is containerized and built with Flask. Please refer to the dedicated README for setup and usage instructions.

---

📦 Tech Stack

Layer	Technology Used
Backend	Spring Boot, MySQL, RabbitMQ, Java
Frontend	React, TypeScript, Vite, TailwindCSS
ML Model	Python, Flask, Scikit-learn, Pandas
Auth/Security	JWT, Refresh Tokens, BCrypt
DevOps	Docker, Docker Compose, GitHub Actions

---

🖼️ Application Screenshots

🧩 Feature	📸 Screenshot
🏠 Home Screen
📝 Signup
🔐 Login
📊 Dashboard
💸 Send Money
📈 Recent Transactions
📋 All Transactions
📉 ML Model – Fraud Transaction Detected
📈 ML Model – Legitimate Transaction Detected
🤖 AI Model – Fraud Transaction Detected
🧾 AI Model – Legitimate Transaction Detected

---

📊 Fraud Detection System

We developed a machine learning model capable of detecting fraudulent transactions using patterns in past data.

✅ Features:

• Binary classification (Fraud vs Legitimate)
• RESTful API in Flask
• Google Gemini AI integration to explain predictions
• Real-time integration with backend for fraud alerts

📘 Fraud Detection README

---

🤝 Contributors

Name	Role
Mohammed Varaliya	Project Lead, Backend, Frontend, ML Model & Integration, UI/UX, System Design
Vraj Shah	Project Lead, Backend Development, Docker, Frontend, UI/UX, System Design
Jayesh Mal	UI/UX, Frontend

---

📎 Useful Links

• 🔗 Backend Repo: java_backend
• 🔗 Frontend Repo: Online-Banking-Application-Frontend
• 🔗 ML Fraud Detection: ML Fraud API
• 🔗 System Design Workspace: Design Docs

---