Keystroke statistical learning model for web authentication

Keystroke typing characteristics is considered as one of the important biometric features that can be used to protect users against malicious attacks. In this paper we propose a statistical model for web authentication with keystroke typing characteristics based on Hidden Markov Model and Gaussian Modeling from Statistical Learning Theory. Our proposed model can substantially enhance the accuracy of the identity authentication by analyzing keystroke timing information of the username and password. Results of the experiments showed that our scheme achieved by far the best error rate of 2.54%.

[1]  Jan H. P. Eloff,et al.  Enhanced Password Authentication through Fuzzy Logic , 1997, IEEE Expert.

[2]  Paulo Sérgio Tenreiro Magalhães,et al.  An improved statistical keystroke dynamics algorithm , 2005 .

[3]  Xian Ke,et al.  Typing patterns: a key to user identification , 2004, IEEE Security & Privacy Magazine.

[4]  Peter Norvig,et al.  Artificial Intelligence: A Modern Approach , 1995 .

[5]  Sajjad Haider,et al.  A multi-technique approach for user identification through keystroke dynamics , 2000, Smc 2000 conference proceedings. 2000 ieee international conference on systems, man and cybernetics. 'cybernetics evolving to systems, humans, organizations, and their complex interactions' (cat. no.0.

[6]  Claudia Picardi,et al.  User authentication through keystroke dynamics , 2002, TSEC.

[7]  Norman Shapiro,et al.  Authentication by Keystroke Timing: Some Preliminary Results , 1980 .

[8]  Aurangzeb Khan,et al.  Enhancing login security using keystroke hardening and keyboard gridding , 2005 .

[9]  Stuart J. Russell,et al.  Artificial Intelligence , 1999 .

[10]  Lawrence R. Rabiner,et al.  A tutorial on hidden Markov models and selected applications in speech recognition , 1989, Proc. IEEE.

[11]  Fabian Monrose,et al.  Authentication via keystroke dynamics , 1997, CCS '97.