Recent Advances in the Development of a Long-Text-Input Keystroke Biometric Authentication System for Arbitrary Text Input

Behavioral biometric verification of student identity in online course assessment and authentication of authors in literary works

“…Nevertheless, there has been a long history of commercially unsuccessful implementations aimed at continuous recognition of a typist. While most previous work dealt with short input (passwords or short name strings) [1,7,14,15,16], some used long free (arbitrary) text input [2,8,11,13,19,20]. Free-text input as the user continues typing allows for continuous authentication [5,12,13,17] which can be important in online exam applications [6,19].…”

Utilizing overt and latent linguistic structure to improve keystroke-based authentication

“…Messerman, et al [18] improves upon the scalability of Gunetti & Picardi [12] to provide authentication in a real-time environment. Monaco, et al [19] also provides comparable results, and importantly also demonstrates the effects of population size and window-sample size on success rate. Finally, Zhong, et al [37] utilizes a novel distance metric that incorporates the best aspects of both Manhattan and Mahalanobis distance, to achieve best-in-class results on the CMU keystroke dynamics benchmark dataset of subjects typing passwords.…”

One-handed Keystroke Biometric Identification Competition

“…The final leaderboard of the competition is shown in Table 2. Baseline results were obtained with a 1-nearestneighbor classifier using scaled Manhattan distance [2] and 218 commonly used keystroke features [12]. This baseline is different than the baseline obtained by the starter code provided to participants.…”

“…There are commercial products available that analyze a sequence of keystrokes for human identification, or provide additional security through password hardening and continuous authentication. It is common to see error rates below 10% for short text authentication [11], and below 1% in long text applications [12]. In terms of continuous authentication, an intruder can accurately be identified in less than 100 keystrokes [4].…”