Pace University's Keystroke Biometric System (PKBS)

In Seidenberg's School of CSIS at Pace University, over the last eight years we have developed robust text-input keystroke biometric systems for identification (one-of-n response) and for authentication (accept/reject response). As far as we can determine, these text-input systems are currently the best in existence and we strongly desire to maintain our leadership in this area. While there are several commercial keystroke biometric systems that operate on passwords (a few seconds of input), our systems operate on arbitrary text input of longer duration, usually in minutes, and can therefore capture more robust statistical-based features. We have also worked in the related computer-user areas of mouse activity, stylomery (forensic authorship), and semantic operational biometrics. We have presented experimental results at several conferences and have recently published a book chapter and journal article.

There are several interesting applications of this work with related funding opportunities. The Department of Defense is interested in intruder detection and their "Active Authentication" project wants to continually authenticate users of all government machines, military and non-military. A second application is verifying the identity of customers making online transactions. A third is verifying the identity of students taking online tests, an application important for the 2008 United States Higher Education Opportunity Act which requires institutions of higher learning to make greater online access control efforts by adopting ubiquitous identification technologies. While all three applications are similar in terms of authenticating the user, faster discovery is required in the first two to prevent significant harm.

In 2013 a major system improvement dropped the error rate by a factor of five over those reported in reference [1]. The improved system results on a population of 120 users yielded a biometric performance rate of over 99% for an input of 100 or more keystrokes (100 keystrokes is about 15 words).

Primary references
  1. John V. Monaco, Ned Bakelman, Sung-Hyuk Cha, and Charles C. Tappert, "Developing a Keystroke Biometric System for Continual Authentication of Computer Users," Proc. 2012 European Intelligence and Security Informatics Conference, Denmark, August 2012. PDF   Slides
  2. John C. Stewart, John V. Monaco, Sung-Hyuk Cha, and Charles C. Tappert, "An Investigation of Keystroke and Stylometry Traits for Authenticating Online Test Takers," Proc. IEEE Int. Joint Conf. Biometrics, Washington D.C., October 2011. PDF
  3. C.C. Tappert, S. Cha, M. Villani, and R.S. Zack, "Keystroke Biometric Identification and Authentication on Long-Text Input," Int. Journal Information Security and Privacy (IJISP), 2010. PDF
  4. R.S. Zack, C.C. Tappert, and S. Cha, "Performance of a Long-Text-Input Keystroke Biometric Authentication System Using an Improved k-Nearest-Neighbor Classification Method," Proc. IEEE 4th Int. Conf. Biometrics, Washington D.C., September 2010. PDF   Slides
  5. C.C. Tappert, M. Villani, and S. Cha, "Keystroke Biometric Identification and Authentication on Long-Text Input," pp 342-367, Chapter 16 in Behavioral Biometrics for Human Identification: Intelligent Applications, Edited by Liang Wang and Xin Geng, Medical Information Science Reference, 2010. PDF
  6. M. Villani, C.C. Tappert, G. Ngo, J. Simone, H. St. Fort, and S. Cha, "Keystroke Biometric Recognition Studies on Long-Text Input under Ideal and Application-Oriented Conditions," Proc. CVPR 2006 Workshop on Biometrics, New York, NY, June 2006. PDF