Improving keystroke dynamics authentication system via multiple feature fusion scheme

This paper reports the performance and effect of diverse keystroke features combination on keystroke dynamic authentication system by using fusion scheme. First of all, four types of keystroke features are acquired from our collected dataset, later then transformed into similarity scores by the use of Gaussian Probability Density Function (GPD) and Direction Similarity Measure (DSM). Next, three fusion schemes are introduced to merge the scores pairing with six fusion rules. Result shows that the finest performance is obtained by the combination of both dwell time and flight time collectively. Finally, this experiment also investigates the effect of using larger dataset on performance, which turns out to be rather consistent.