Fraud detection

Payment transactions are a hacker’s favourite target. Thus, there is all the more need, for intelligent protective measures to ward off professional fraudsters. Banks tend to use rule-based systems, which are intended to detect fraud attempts. The disadvantage of this kind of system is the lack of flexibility in the event of new attack scenarios. Which in turn results in rising costs, increased personnel expenditure and dissatisfied customers due to excess false positives.


In a pilot project with one customer, Ergon has developed a procedure to address this exact issue: an intelligent fraud detection system that makes use of machine learning. Initial results show the self-learning system is better at identifying fraud attempts than human experts. The concept is now being developed further.


However, In order to provide comprehensive protection, fraud detection should be used in combination with other methods. Ergon will be offering this system in combination with Airlock Suite thereby enabling ‘prevention’ and ‘detection’ mechanisms to work in tandem.