False positive
A false positive in fraud monitoring is an alert raised on a transaction that turns out to be legitimate. Each one costs analyst time and can inconvenience a genuine customer, and false positives trade off directly against missed fraud, which is the false negative.
In more detail
False positives and false negatives pull against each other. Tightening a rule to catch more fraud usually raises false positives; loosening it to cut analyst workload usually lets more fraud through. Neither can be driven to zero, and a programme claiming zero false positives has almost certainly set its rules too loosely.
The useful discipline is to measure false positives per rule rather than in total. An overall rate hides a single noisy rule generating most of the wasted work, while a per-rule view shows which rule to tighten, retire or rewrite.
A reporting layer records the outcome of every alert and matches it to the rule that raised it, so the noisy rule can be identified.
Example
A rule flags every transfer above a fixed amount. Over a month, analysts clear nearly all of its alerts as genuine payroll and supplier payments. Reporting by rule shows this one rule consuming most of the queue, and it is rewritten to exclude established payees.
Related capability
False-positive reporting on the platform
Related terms
Discuss your transaction monitoring requirements
Tell us which transaction flows you need to monitor, which systems Akoben would integrate with, and how your fraud team works today.
Request a demo