Danske Bank deploys BehavioSec’s Behavioral Biometrics Technology
Danske Bank is well known for providing their customers with great user experience. In their effort to further constantly challenge and increase user experience, they began moving into easy-to-handle strong security unified across platforms and devices.
In doing so, previous effective methods for fraud detection had to be abandoned, and the wish to bind sessions to customers rather than devices became apparent. They evaluated the BehavioWeb fraud detection and chose BehavioSec as the behavioral biometrics partner most likely to succeed in ensuring that online transactions originate from the correct user every time.
Moving into the spacey area of biometrics, the most important qualities – besides a well functioning product – is the willingness to interact, options to tweak and develop on an ongoing basis, and a solid foundation in proving proper operation. With this choice, we feel well positioned for our future security developments.
– Ebbe Skak Larsen, Chief Security Architect Danske Bank
Below you will see how we structured our pilot with Danske Bank, using a data mining life cycle. The functionality was presented in a report to get direct feedback on matching and verification of a subset of users. The trial included 18,000 live users and 540,000 transactions, and the goal of the trial was to show our performance and accuracy on real users. Danske Bank was provided with user statistics for FAR/FRR/EER, so that this technology could be compared to other alternatives on the market. Once their internal analysis wasperformed a decision was made to move forward with field testing on real users.
In our pilot trial with Danske Bank, our BehavioWeb product was able to properly distinguish between correct user and imposter with stolen credentials in 99.7% of the cases during, using BehavioSec patented continuous authentication for Danske Banks E-Banking platform (see whitepaper for details).
These results led Danske Bank to deploy BehavioWeb on real users in live environment to further field test the technology. Full deployment to 2.5M users is expected during 2014