Mouse Movement Biometric

Background

Mouse motion behavior is believed to be unique to an individual. The way a user moves a mouse in an application depends on hand shape and size, muscle control, and experience with the system. Mouse motion is ubiquitous for desktop computer users so there is much motivation to be able to identify and verify a user based on mouse movements.

We have collected mouse motion data from users taking exams on Moodle for several semesters. We currently have over 50 users who have taken 10 quizzes each.

Project

The goal of this project is to determine how the mouse motion data can be used to differentiate between users. This may be performed by first extracting features from the mouse motion tracks, such as the speed and accuracy of motion between certain elements on a webpage.

One of the major obstacles in validating any approach is that the mouse motion data has been collected from a variety of devices and operating systems. The OS artificially speeds up the motion of the pointer, so what we observe is actually not the motion of the mouse itself. The first step in this project would be to reverse engineer the artificial acceleration of the pointer motion to retrieve the actual mouse motion. See this paper on pointer ballistics in Windows XP (which is now only available through an archive): Pointer ballistics in Windows XP.

This project will address this critical obstacle that must be solved to continue research is this area. A suggested direction for this project is through the use of a Kalman filter, see Kalman filter python library. For your python environment, the Anaconda distribution comes with all the popular scientific libraries, see downloads.

Necessary Skills Required

Work on this project requires the following skills: data analysis, programming in a scientific language (matlab, python, R, octave, julia, etc.), and decent math background. This is a difficult project, and without these skills the learning curve may inhibit real progress. Some familiarity with operating systems would also be helpful.

References

References to previous work in the mouse movement area are listed below but are not really relevant to the work of this project.
  1. Francis Buckley, Vito Barnes, Thomas Corum, Stephen Gelardi, Keith Rainsford, Phil Dressner, and Vinnie Monaco, Team 5, Fall 2014 Technical Report.
  2. Hedieh Zandikarimi, Frank Lin, Celia Carlos, Justin Correa, Phil Dressner, and Vinnie Monaco, Design of a Mouse Movement Biometric System to Verify the Identity of Students Taking Multiple-Choice Online Tests, Proc. Research Day, CSIS, Pace University, May 2014.
  3. Francisco Betances, Adam Pine, Gerald Thompson, Hedieh Zandikarimi, and Vinnie Monaco, Mouse Biometric Authentication, Proc. Research Day, CSIS, Pace University, May 2014.
  4. Pedro Xavier de Oliveira, Venugopala Channarayappa, Eamonn O'Donnel, Bappaditya Sinha, Aswinkumar Vadakkencherry, Tushar Londhe, Umesh Gatkal, Ned Bakelman, John V. Monaco, and Charles C. Tappert, Mouse Movement Biometric System, Proc. Research Day, CSIS, Pace University, 2013. slides
  5. Chao Shen, et al., User Authentication Through Mouse Dynamics, IEEE Trans. Info. Forensics and Security, 8-1, Jan 2013.
  6. Allen Newell, Section on Fitts' Law, Unified Theories of Cognition (The William James Lectures), Harvard University Press, 1994.
  7. Nkem Ajufor, Antony Amalraj, Rafael Diaz, Mohammed Islam, Michael Lampe, Refinement of a Mouse Movement Biometric System, Proc. Research Day, CSIS, Pace University, 2007.
  8. Nan Zheng, Aaron Paloski, and Haining Wang, An Efficient User Verification System via Mouse Movements, CCS’11, Chicago, October 2011.
  9. Maja Pusara and Carla E. Brodley, User Re-Authentication via Mouse Movements, VizSEC/DMSEC'04, Washington DC, 2004.