I saw a need to create a single source of information on payment fraud at Tradition Capital Bank. Payment fraud impacts all divisions and is a potential threat to all clients. A single source of information can help people determine where to focus anti-fraud efforts. My objectives included collecting data around fraud, creating visualizations, and exploring relationships between clients and transactions.  

I cleaned and stored data on payment fraud in our data warehouse. This involved creating an ETL process to continually update the warehouse. I also created a Nintex form to track cases of debit card fraud because of the strict data requirements for these cases. By storing this data in the warehouse, I was able to create reports and visualizations to distribute to employees via their dashboards. Additionally, I used clustering to explore Tradition Capital Bank’s business clients and for the exploration of transactions on a single account. Future research includes using different data sources and model types for predicting payment fraud. Fraudsters will improve their methods in order to be more successful at committing fraud, and data scientists need to be prepared to improve their models and analyses in order to minimize their losses.