User authentication through keystroke dynamics

Abstract

Unlike other access control systems based on biometric features, keystroke analysis has not led to techniques providing an acceptable level of accuracy. The reason is probably the intrinsic variability of typing dynamics, versus other---very stable---biometric characteristics, such as face or fingerprint patterns. In this paper we present an original measure for keystroke dynamics that limits the instability of this biometric feature. We have tested our approach on 154 individuals, achieving a False Alarm Rate of about 4% and an Impostor Pass Rate of less than 0.01%. This performance is reached using the same sampling text for all the individuals, allowing typing errors, without any specific tailoring of the authentication system with respect to the available set of typing samples and users, and collecting the samples over a 28.8-Kbaud remote modem connection.

User identity verification via mouse dynamics

“…An FAR 0.43% and an FRR of 1.75% were achieved. [3], extracted the typing durations of two (di-graph) and three (tri-graph) consecutive characters from a sample and used to associate it to a user. The extracted graphs were ordered by their duration and their relative ordering was compared to the relative order of the training samples of other users.…”

“…Gunetti et al [24] extended the approach of [3] to also handle free text. Furthermore, they proposed another distance measure based on absolute times.…”

Keystroke Dynamics Overview

“…trigraph are heavily used in (Bergadano et al, 2002). de Ru & Eloff (1997) use a concept of typing difficulty based on the fact that certain key combinations are more difficult to type than other.…”

“…We have seen that keystroke dynamics allows to secure the authentication process by verifying the way of typing the credentials. It can also be used to secure the session after its opening by detecting the changing of typing behavior in the session (Bergadano et al, 2002;Marsters, 2009). In this case, we talk about continuous authentication (Rao, 2005), the computer knows how the user interacts with its keyboard.…”

Cybermetrics: User Identification through Network Flow Analysis

“…The idea of user recognition and identification by ex-ploitation of biometric patterns has long been known [1], [2]. More recent studies look at dynamics of certain actions performed by the user -be it an on-line game-play activity [3], which shows that the idle and active times in the game are representative of the user; a keystroke analysis [4], which provides an impressive 96% correctness rate at user differentiation; or user-mouse interaction dynamics [5], establishing a behavioral characteristic that can be used as an additional security feature. The second area of research uses passive network traffic monitoring techniques for performance analysis, application type/protocol identification, anomaly and intrusion detections.…”