Pioneered by BehavioSec, Behavioral Biometrics uses characteristics of human behavior to authenticate individuals based on how they perform digital inputs, such as mouse movements, typing rhythm, touch and swipe gestures, or the particular way an individual holds their device. Behavioral biometrics provides continuous authentication to verify an individual by monitoring known biometric behavior in the background without negatively impacting their experience.
BehavioSec introduced this technology in 2007. Today, BehavioSec serves a wide range of Global 2000 enterprises while continuing to work on advanced Research and Development projects, including projects for the U.S. Department of Defense’s Defense Advanced Research Projects Agency (DARPA).
BehavioSec’s patented approach enables enterprises to seamlessly integrate our Behavioral Biometrics software platform into their existing and future mobile and web apps.
Using a rich suite of APIs and lightweight SDKs, BehavioSec’s software provides enterprises with realtime actionable intelligence scores and continuous authentication of users, preventing account takeovers, new account fraud and machine automated attacks.
BehavioSec has been able to verify digital identities with 99.7% accuracy on millions of users. As Behavioral Biometrics is dependent on user behavior, BehavioSec’s software gets better as it collects more data. At one of BehavioSec’s long running clients where behavioral data is limited to only a 6 digit mobile PIN-code, BehavioSec continues to deliver a digital identity accuracy of 97.4%. See for yourself by requesting a demo at https://www.behaviosec.com/demos/
A threshold value is used to separate an imposter or fraudster from a true user. It directly influences the False Acceptance Rate (FAR) and the False Rejection Rate (FRR). When a score is above the threshold, we consider the behavior to belong to the correct user. If the score is below the threshold, we consider it to belong to the imposter.
The EER is a measurement of accuracy in biometric systems used to predetermine the threshold values for its False Acceptance Rate (FAR) and False Rejection Rate (FRR) . A single FAR without its corresponding FRR does not gauge accuracy since it is possible that a system with the lowest FAR has a high FRR. Whereas the EER is found at the threshold level where the False Acceptance Rate (FAR) and False Rejection Rate (FRR) are equal. The lower the EER, the more accurate the system is considered to be.
There is no significant difference in accuracy between web and mobile channels. While mobile has more sensor modalities, people tend to type more rhythmically on desktops. No matter the channel, BehavioSec is able to verify a digital identity with an accuracy in the high 90’s.
BehavioSec is live on tens of millions of users worldwide. Each instance can be easily scaled with load balancers to handle any number of requests per second. Neither the number of users nor peak load impact the accuracy of a system.
BehavioSec’s software is easy to set up and scale across countries and continents. Several of BehavioSec’s Global 2000 customers have integrated BehavioSec’s Behavioral Biometrics software on global applications, with no negative impact on latency or accuracy.
While unlikely, it is certainly possible for two or more users to have similar behavioral profiles. However, that’s not a problem for BehavioSec’s Behavioral Biometrics software. BehavioSec’s software verifies users individually against their own past behavior rather than against other users.
Not at all. BehavioSec’s software isn’t affected by the battery level of the device. Furthermore, data collection on-device is extremely light weight and only active in-app, resulting in a negligible impact on the device’s energy consumption.
No, BehavioSec has supported many diverse hosting topologies across a wide range of Global 2000 customers without experiencing any significant differences in our impressive accuracy or performance.