This project aims to detect fraudulent transactions made by European credit cardholders in September 2013. The dataset contains 284,807 transactions, out of which 492 are fraudulent. The data is highly imbalanced, with the positive class (fraudulent transactions) accounting for only 0.172% of all transactions.
The dataset used for this project contains both real and fake credit card numbers. The data is imbalanced, with a disproportionate number of fake credit card numbers compared to real credit card numbers. To address this imbalance, the data is oversampled and undersampled before training the model.