Machine Learning & Behavioral Biometrics

Data scientists have discovered an interesting fact: People operate in unpredictable ways. There is no generic model that everyone can be measured against, generating results that are accurate enough both on the false positive side, which is critical for preventing fraud and identity theft -- and on the false negative side, which is important for user experience. Today’s behavioral biometric technologies can capture more than 2,000 parameters from a mobile device, including the way a person holds the device, scrolls, toggles, finger pressure and how they respond to stimuli from applications. Machine learning is used in all facets of behavioral biometrics; and this is what makes this technology magical.

By applying machine learning to behavioral biometrics, we are able to learn from human behavior and continuously improve user profiles that can be used to validate sessions or transactions. As a person exhibits familiar behavior patterns, their behavior can be forecasted with increasingly accurate confidence intervals. Machine learning helps to break through the clutter of the various signals and find the consistencies in the behavioral patterns over time, even with the changes and adaptations in human behavior patterns.

Nakul Munjal is the founder and CEO of Status Identity; a Herndon, VA based startup that specializes in applying behavioral biometrics to multi-factor authentication. His background includes over a decade in identity and access management space with IBM and Micro Focus.

For our second speaker, we will have Hans Bergman from BehavioSec speaking about some of the use cases they have solved through their biometric solutions.

Behavioral biometrics technology verifies end users based on their interaction with their devices. We have deployed our technology on varies use-cases; government projects, e-commerce, and retail banking. Our talk will focus on use-cases for behavioral biometrics which are in production today and explore use-cases of near and far future.