Credit Card Fraud Detection

This project is a machine learning model for detecting credit card fraud. It uses a RandomForestClassifier, which was found to be the best model for this problem statement.

Dataset

The dataset used in this project is imbalanced, with the majority of credit card transactions being normal and a very small percentage being fraudulent. To handle this imbalance, the Synthetic Minority Over-sampling Technique (SMOTE) is used to oversample the minority class (fraudulent transactions).

Bulk Prediction

The project is designed to make predictions in bulk. It accepts a file of inputs and outputs a file of predictions.

Running the Project

To run the project, follow these steps:

1. Clone the repository.
2. Install the required dependencies.
3. Run the Flask application.

git clone https://github.com/aqib0770/Credit_Card_Fault_Detection.git
pip install -r requirements.txt
python app.py