Fraud Detection with Imbalanced Data in Money Laundering Dataset

This project explores methods to handle imbalanced data using a money laundering dataset. The issue is that there are many more normal transactions compared to fraudulent ones. To address this, the focus is on upsampling, which involves creating more examples of fraud cases to balance the data. Different models like Random Forest, Gradient Boosting, and Logistic Regression are used to predict fraud. The results are evaluated using precision, recall, and F1-score to measure how well the models can detect fraudulent transactions.