PROJECT TITLE :

Person Identification by Keystroke Dynamics using Pairwise User Coupling - 2017

ABSTRACT:

Thanks to the increasing vulnerabilities in cyberspace, security alone isn't enough to prevent a breach, but cyber forensics or cyber intelligence is also needed to stop future attacks or to spot the potential attacker. The unobtrusive and covert nature of biometric information assortment of keystroke dynamics has a high potential for use in cyber forensics or cyber intelligence. During this paper, we investigate the usefulness of keystroke dynamics to ascertain the person identity. We propose 3 schemes for identifying an individual when typing on a keyboard. We use numerous Machine Learning algorithms in combination with the proposed pairwise user coupling technique and show the performance of every separate technique still as the performance when combining two or additional together. In particular, we show that pairwise user coupling during a bottom-up tree structure scheme provides the best performance, each regarding accuracy and time complexity. The proposed techniques are validated by using keystroke knowledge. But, these techniques could equally preferably be applied to alternative pattern identification issues. We have a tendency to have additionally investigated the optimized feature set for person identification by using keystroke dynamics. Finally, we tend to conjointly examined the performance of the identification system when a user, not like his traditional behaviour, varieties with only one hand, and we show that performance then is not optimal, as was to be expected.