One factor impeding the acceptance of biometric security is the lack of actual classification rates based on usage in a real world setting. While previous results from controlled lab experiments have led to significant advancements in the field, they cannot be generalized nor do they accurately provide implications for practical deployment. This research investigates classification rates that can be achieved using a keystroke analysis prototype in an uncontrolled setting. We introduce a new pressure related feature, vpdelta, for classifying typing patterns and conduct a 5 week long field experiment to examine the variability of typing patterns over time. A classification rate of 87% was obtained during actual use and little variability in typing patterns was observed during the experiment. This finding helps to validate and extend previous lab studies, and provides increased support for the application of keystroke analysis in a real world environment.

Bleha, S., Knopp, J., and Obaidat, M. (1992). Performance of the perceptron algorithm for the classification of computer users. Paper presented at the Proceedings of the 1992 ACM/SIGAPP Symposium on Applied Computing: technological challenges of the 1990's.

Bleha, S. A. and Obaidat, M. S. (1991). Dimensionality reduction and feature extraction applications in identifying computer users. Systems, Man and Cybernetics, IEEE Transactions on, 21(2), 452-456.

Chandra, A. and Calderon, T. (2005). Challenges and constraints to the diffusion of biometrics in information systems. Communications of the ACM, 48(12), 101-106.

Eltahir, W., Salami, M. J. E., Ismail, A., and Lai, W. (2008). Design and evaluation of a pressure-based typing biometric authentication system. EURASIP Journal on Information Security, 2008(1), 345047.

Kotani, K. and Horii, K. (2005). Evaluation on a keystroke authentication system by keying force incorporated with temporal characteristics of keystroke dynamics. Behaviour & Information Technology, 24(4), 289-302.

Leberknight, C. (2015). An Embedded System for Extracting Keystroke Patterns Using Pressure Sensors (in press). International Journal of Biometrics.

Leggett, J., Williams, G., Usnick, M., and Longnecker, M. (1991). Dynamic identity verification via keystroke characteristics. International Journal of Man-Machine Studies, 35(6), 859-870.

Lin, D.-T. (1997). Computer-access authentication with neural network based keystroke identity verification. Paper presented at the Neural Networks, 1997. International Conference on Neural Networks, IEEE.

Loy, C. C., Lai, W., and Lim, C. (2005). Development of a pressure-based typing biometrics user authentication system. ASEAN Virtual Instrumentation Applications Contest Submission.

Loy, C. C., Lai, W. K., and Lim, C. P. (2007). ‘Keystroke patterns classification using the ARTMAP-FD neural network’, Third International Conference on Intelligent Information Hiding and Multimedia Signal Processing, 2007. IIHMSP 2007, November, IEEE, Vol. 1, pp.61–64.

Nonaka, H. and Kurihara, M. (2004). Sensing Pressure for Authentication System Using Keystroke Dynamics. Paper presented at the International Conference on Computational Intelligence.

Robinson, J. A., Liang, V., Chambers, J. M., and Mackenzie, C. L. (1998). Computer user verification using login string keystroke dynamics. Systems, Man and Cybernetics, Part A: Systems and Humans, IEEE Transactions on, 28(2), 236-241.

Siponen, M. T. and Oinas-Kukkonen, H. (2007). A review of information security issues and respective research contributions. ACM Sigmis Database, 38(1), 60-80.

Umphress, D. and Williams, G. (1985). Identity verification through keyboard characteristics. International Journal of Man-Machine Studies, 23(3), 263-273.