User identification of numerical keypad typing patterns with subtractive clustering fuzzy inference

Password security of user accounts is a major concern in today's fast moving and advanced computing world with password authentication the most commonly used identification mechanism. Its security can be enhanced using typing biometrics as a transparent layer of user authentication. This paper describes a password hardening mechanism that is able to recognize different users based on their time latencies. Our research focuses on only using the time period between keystrokes as the measure of the individual's typing pattern. Seven individuals used the same 7-digit password which allowed the system to record the time interval between each digit. Identification improvements up to 3% were achieved with Subtractive Clustering Fuzzy Inference when rule conflicts were removed.