Fraud Detection Machine Learning Model

The goal of this project was to create a ML model with at least 70% recall. When working with E-Commerce Transactions, recall is vital. The recall demonstrates the amount of fraudulent charges the model is finding.

While the recall is just above the goal, the precision is low. I would suggest further focus on improving the precision, so we can minimize the amount of false red flags for fraud.

I also found the top 10 factors that are related to fraud.

Thank you to Shriyash Jagtap for providing the dataset!

Dataset Resource: Jagtap, Shriyash. (2024, April 7). Fraudulent E-Commerce Transactions. Kaggle. https://www.kaggle.com/datasets/shriyashjagtap/fraudulent-e-commerce-transactions