FinGuard --- Credit Card Fraud Detection System

FinGuard is an end-to-end fraud detection system built using Flask, scikit-learn, SHAP explainability, and Plotly visual analytics. This project demonstrates real-world AI engineering skills with clean UI, API documentation, and reproducible ML workflows.

---

Quick Demo

Run the app locally and open:

http://127.0.0.1:5000/

---

What This Project Contains

• Flask application (app/)
• SHAP explainability UI
• Plotly ROC visualization
• Swagger API documentation (/docs)
• Screenshot assets (docs/images/)
• Clean project structure suitable for recruiters

---

Tech Stack

• Python 3.13\
• Flask\
• scikit-learn\
• SHAP\
• Plotly\
• HTML/CSS/JS (modern UI)

---

How It Works

1. Load dataset (creditcard-database.csv)
2. Train Logistic Regression baseline model
3. Save model + ROC metrics + SHAP explanations
4. Provide prediction form + API + visualizations

---

Run Locally

git clone https://github.com/tonumayworkspace-creator/FinGuard.git
cd FinGuard
python -m venv venv
venv\Scripts\activate
pip install -r requirements.txt
pip install shap plotly matplotlib
python app\app.py

---

API Documentation (Swagger)

http://127.0.0.1:5000/docs

Example Request:

{
  "features": {
    "Time": 34500,
    "Amount": 120.5,
    "V1": -1.23
  }
}

---

Screenshots

(Add matching images in docs/images/)

---

Author

Tonumay Bhattacharya
Data Science & AI Engineering Enthusiast

GitHub: https://github.com/tonumayworkspace-creator\ LinkedIn: (add link here)

---

License

MIT License