FinGuard — Smart Financial Crime Detection

A full-stack ML-powered financial fraud detection application built with React + FastAPI.

---

Project Structure

internship.proj/
├── backend/
│   ├── main.py              # FastAPI server + ML model
│   └── requirements.txt
└── frontend/
    ├── src/
    │   ├── components/      # Sidebar, StatCard, TransactionRow
    │   ├── pages/           # Dashboard, FraudAnalyzer, Transactions, Analytics, Alerts, Settings
    │   ├── utils/api.js     # Axios API calls
    │   ├── App.jsx
    │   ├── main.jsx
    │   └── index.css
    ├── index.html
    ├── package.json
    ├── vite.config.js
    └── tailwind.config.js

---

Quick Start

1. Start the Backend

cd backend
pip install -r requirements.txt
uvicorn main:app --reload --port 8000

Backend runs at: http://localhost:8000 API Docs at: http://localhost:8000/docs

2. Start the Frontend

cd frontend
npm install
npm run dev

Frontend runs at: http://localhost:3000

---

Features

• Dashboard — Live overview with stats, charts, recent transactions
• Live Transaction Feed — Real-time stream with risk scores and filters
• Fraud Analyzer — Submit any transaction for instant ML risk scoring
• Analytics — Model performance metrics, confusion matrix, charts
• Alert Center — Security alerts with severity levels and resolution
• Settings — Configure detection thresholds, model, and notifications

Tech Stack

Layer	Technology
Frontend	React 18, Vite, Tailwind CSS, Recharts, Framer Motion
Backend	FastAPI, Python
ML	scikit-learn, RandomForest, SMOTE, XGBoost
API	REST (Axios)
Deploy	Docker-ready

API Endpoints

Method	Endpoint	Description
GET	/api/analytics/overview	System overview stats
GET	/api/metrics	ML model metrics
GET	/api/transactions/stream	Live transactions
GET	/api/analytics/chart-data	Charts data
POST	/api/predict	Predict fraud risk
POST	/api/train	Retrain model