Performance Enhancement on Keystroke Dynamics by Using Fusion Rules
Keystroke dynamics refers to the timing information that expresses precisely when each key was pressed and released as a person types. In this paper, we present a novel keystroke dynamic recognition system by using a novel fusion approach. Firstly, we extract four types of keystroke latency as the feature data from our dataset. We then calculate their mean and standard deviation to be stored as template. The test feature data will be transformed into similarity scores via Gaussian Probability Density Function (GPD). We also propose a new technique, known as Direction Similarity Measure (DSM), to measure the trend differential among each digraph in a phrase. Lastly, various fusion rules are applied to improve the final result by fusing the scores produced by GPD and DSM. Best result with equal error rate of 2.791% is obtained when the AND voting rule is used.