Keystroke Patterns Classification Using the ARTMAP-FD Neural Network

This paper presents the development of a keystroke dynamics-based user authentication system using the ARTMAP-FD neural network. The effectiveness of ARTMAP- FD in classifying keystroke patterns is analyzed and compared against a number of widely used machine learning systems. The results show that ARTMAP-FD performs well against many of its counterparts in keystroke patterns classification. Apart from that, instead of using the conventional typing timing characteristics, the applicability of typing pressure to ascertaining user´s identity is investigated. The experimental results show that combining both latency and pressure patterns can improve the equal error rate (ERR) of the system.