Machine learning and artificial intelligence (AI) have a long track record of successful use in payment fraud. However, the hype can get overwhelming and it can get difficult for many to distinguish myth from reality. Machine learning can clearly distinguish legitimate and fraudulent behaviours if done properly while adapting over time to new, previously unseen fraud tactics.
Protecting customers’ accounts is a complex and dynamic process and also a challenge where machine learning thrives. Fraud detection professionals should consider adaptive technologies designed to sharpen responses for continual performance improvement, particularly on marginal decisions. Either just above or just below the cutoff, these are the transactions that are very close to the investigative triggers. There is a fine line between a false positive event — a legitimate transaction which has scored too high, and a false negative event — a fraudulent transaction which has scored too low, and it is on these margins where accuracy is most critical. With up-to-date knowledge of the threat vectors an institution is facing, adaptive analytics sharpen this distinction.
Adaptive analytics technologies result in a more precise separation between frauds and non-frauds by automatically adapting to recently confirmed case disposition to improve sensitivity to shifting fraud patterns. When an analyst investigates a transaction, the outcome is fed back into the system to accurately reflect the fraud environment that analysts are facing, including new tactics and subtle fraud patterns that have been dormant for some time. This adaptive modelling technique automatically modifies the weights of predictive features within the underlying fraud models and hence is a powerful tool that improves fraud detection performance on the margins and stops new types of fraud attacks.