Statistical fusion approach on keystroke dynamics

Keystroke dynamics refers to a user's habitual typing characteristics. These typing characteristics are believed to be unique among large populations. In this paper, we present a novel keystroke dynamic recognition system by using a fusion method. Firstly, we record the dwell time and the flight time as the feature data. We then calculate their mean and standard deviation values and stored. The test feature data will be transformed into the scores via Gaussian probability density function. On the other hand, we also propose a new technique, known as Direction Similarity Measure (DSM) to measure the differential of sign among each coupled characters in a phrase. Lastly, a weighted sum rule is applied by fusing the Gaussian scores and the DSM to enhance the final result. The best result of equal error rate 6.36% is obtained by using our home-made dataset.