Alaric offers Bayesian alternative to neural card fraud detection

London-based Alaric has launched Fractals, a payment card fraud detection solution which uses 'Bayesian' probability theory to uncover credit card misuse.

  0 Be the first to comment

Alaric offers Bayesian alternative to neural card fraud detection

Editorial

This content has been selected, created and edited by the Finextra editorial team based upon its relevance and interest to our community.

Alaric says the use of Bayesian methods offers significant advantages over neural network-based systems: Fractals models may be computed in a matter of minutes, thus enabling them to be adaptive and to keep pace with changing fraud patterns, as a result of which elapsed time and costs are radically reduced; and it can operate on data from a much smaller number of cardholders, putting the technology in reach of institutions with small to mid-sized cardholder bases.

Mike Alford, Alaric's managing director, adds: "The unique Bayesian approach adopted for Fractals enables fraud officers to interpret the system's reasoning behind a given fraud alert easily and quickly, in contrast to the sometimes impenetrable black box approach associated with neural networks."

Fractals can be delivered as an in-house solution or via Alaric's ASP service.

London-based Alaric was re-launched in 2001 with a cash injection by Northern Venture Managers.

Sponsored [New Impact Study] Catering to a new generation though unified card programmes

Comments: (0)

[Webinar] 2025 Fraud Trends: Synthetic Identity, AI and Incoming MandatesFinextra Promoted[Webinar] 2025 Fraud Trends: Synthetic Identity, AI and Incoming Mandates