An intelligent payment card fraud detection system

Payment cards offer a simple and convenient method for making purchases. Owing to the increase in usage of payment cards, especially online purchases, fraud cases are on rise. The rise creates financial risk uncertainty, as commercial sector, it incurs billions losses each year. However, real transaction records that can facilitate development effective predictive models detection difficult obtain, mainly because issues related confidentially customer information. In this paper, we apply total 13 statistical machine learning card using both publicly available records. results from original features aggregated analyzed compared. A hypothesis test is conducted evaluate whether identified by genetic algorithm better discriminative power, compared with features, detection. outcomes positively ascertain effectiveness undertaking real-world problems.