Credit Card Fraud Detection in Card-Not-Present Transactions: Where to Invest?

Online shopping, already on a steady rise, was propelled even further with the advent of COVID-19 pandemic. Of course, credit cards are dominant way doing business online. The card fraud detection problem has become relevant more than ever as losses due to accumulate. Most research this topic takes an isolated, focused view problem, typically concentrating tuning data mining models. We noticed significant gap between academic findings and rightfully conservative businesses, which careful when adopting new, especially black-box, In paper, we took broader perspective considered from both angle: detected challenges in such feature engineering unbalanced datasets distinguished less lucrative areas invest upgrading systems. Our based real-world CNP (card not present) transactions, type transactions. Data were provided by our industrial partner, international card-processing company. tested different models approaches outlined compared them their existing production systems trace cost-effective system upgrade path.