Using Visualization to detect card-present ATM Fraud

Details of this work are available on request.
This paper was presented at the following conferences.
– 45th Hawaii International Conference on System Sciences (HICSS) in 2011
– American Academy of Forensic Sciences (AAFS) in 2012

Mitigating card-present fraud can be easy when the behaviours associated with the actions have been identified. The situation is much more challenging when you are not sure what you are looking for. The human mind has a powerful ability to identify visual anomalies in large datasets. As part of a defense-in-depth strategy, this can contribute to the evolution and refinement of rule sets that facilitate the detection of the crime prior to the cash-out phase of the illegal operation. This paper investigates the application of visualization combined with human abductive reasoning, to the problem of identifying some behavioural characteristics of card-present fraud. It then demonstrates the behavioural characteristics in context, and demonstrates how this knowledge can be used to guide the evolution of analytical tools to help protect our digital assets.

Leave a Reply

Up ↑

%d bloggers like this: