Credit Card Fraud Detection Web App

An end-to-end machine learning project that detects fraudulent credit card transactions using LightGBM.
This project demonstrates the complete ML workflow — from data preprocessing and model training to saving models (.pkl and .joblib) and deploying with a Flask web app.

---

Features

• Handles large-scale credit card transaction datasets efficiently.
• Uses LightGBM Classifier for high performance.
• Saves trained models as both .pkl and .joblib for flexibility.
• Web UI built with Flask for real-time fraud detection.
• Modular and production-ready project structure.

---

Project Structure

credit-card-fraud-detection/ │ ├── notebooks/ │ └── train_model.ipynb # Training & evaluation notebook │ ├── models/ │ ├── fraud_model.pkl # Saved model (pickle) │ └── fraud_model.joblib # Saved model (joblib) │ ├── app.py # Flask web app for deployment ├── requirements.txt # Dependencies ├── README.md # Project documentation └── data/ └── creditcard.csv # Dataset (not uploaded due to size)

---

Installation

1. Clone the repo:

   git clone https://github.com/your-username/credit-card-fraud-detection.git
   cd credit-card-fraud-detection
   

python -m venv venv source venv/bin/activate # On Windows: venv\Scripts\activate pip install -r requirements.txt

Model training

jupyter notebook notebooks/train_model.ipynb

Start the flask app:

python app.py