Analyzing Impaired-User Input Scenarios for Keystroke Biometric Authentication
In a world where data security is becoming ever more important, the methods we have used to secure our systems are proving to be no longer effective. Passwords and tokens can be easily stolen by a determined hacker and allow unauthorized access to sensitive data. Biometrics has become an ideal option as access is granted to an individual based on a personal attribute as opposed to a code or item that can be obtained. Physiological biometrics, which deals with fingerprints, iris scans, and hand geometry has been widely adopted due to their high accuracy rates; however, there has been evidence suggesting that imposters can recreate a fake fingerprint or other biometric features. In contrast, behavioral biometrics authenticates users by the manner in which they perform a certain task. Some examples are gait (the manner in which we walk), speech, handwriting, and keystroke dynamics. Keystroke Biometrics has been gaining ground as a reliable method for authentication; however, adoption by industry has been sluggish when compared to other biometric systems. To help increase adoption, a more robust system must be developed which takes into account variable input scenarios that occur naturally during a user session. The study analyzes impaired-user input scenarios for keystroke biometric authentication. The study collects keystrokes from 81 students entering various arbitrary long-text responses to quiz questions. It analyzes both hands, left hand only and right hand only scenarios to determine the optimal method to authenticate users across multiple distracted or impaired input scenarios. One drawback of behavioral biometrics that is particularly evident in keystroke dynamics is user-input variability. The findings of this study will strengthen current keystroke biometric systems by allowing the development of methods for the handling of impaired or distracted users. The study also provides a novel approach to better authenticate users of a keystroke biometric authentication system under non-traditional input conditions.
Perez, Gonzalo E, "Analyzing Impaired-User Input Scenarios for Keystroke Biometric Authentication" (2015). ETD Collection for Pace University. AAI3733917.
Remote User: Click Here to Login (must have Pace University remote login ID and password. Once logged in, click on the View More link above)