TypingDNA has published a short primer that explains how its typing biometrics technology works in practice. In broad strokes, the company’s solution analyzes typing patterns to build a unique biometric profile for each user based on the way they interact with a machine.
That profile is based on factors like typing speed and duration. For example, press time refers to the amount of time that a key is pressed, while seek time refers to the length of time that passes between each new keystroke.
However, the full profile is made up of several distinct elements. In that regard, TypingDNA breaks a typing pattern down into a header and the body. The header includes general information about the system and the device being used. It covers everything from the type of device – whether the person is typing on a desktop or a smartphone – to the browser to the operating system, and notes which version of the TypingDNA recorder is being used for capture.
The body, meanwhile, covers the actual keystrokes. It also indicates whether the person is using an internal or an external keyboard, and whether the input is physical or touch-based.
Those headers and bodies are used to put together three different kinds of typing patterns. A Sametext typing pattern is used when a person is asked to type in the exact same phrase or sequence that they used to generate the pattern in the first place, as they would when asked to enter a password, an email address, or a credit card number. Sametext patterns are well-suited to authentication, and add an extra layer of security to traditional login procedures.
Anytext patterns, on the other hand, are used when the new text differs from the text that was entered during the enrollment process, and allow TypingDNA to verify the identity of a user while they are composing an email or having a conversation. An Extended typing pattern supplements that with additional information about the text itself.
TypingDNA’s patterns will pull extra data points from the gyroscope and accelerometer when the individual is typing on a mobile device. The company recently released a two-factor TypingDNA Verify solution that replaces SMS passcodes with typing biometrics.
May 4, 2021 – by Eric Weiss