HOW SIGNIFICANT IS THE INDIVIDUAL DIFFERENCES IN FINGER BASED GESTURES ON SMARTPHONES' TOUCH-SCREEN FOR USER IDENTIFICATION?

Suleyman Al-Showarah
2017 unpublished
This paper analyses the variations in the identification power among set of four features, namely Signature Precision (SP), Finger Pressure (FP), Movement Time (MT), and Speed that were extracted from finger gesture of touch-screens. The differences across users were studied using the above features individually and combined for the purpose of user identification based on the Euclidean distance and the k-nearest neighbour classifier. The paper investigates the performance variations across
more » ... iations across users using a dataset of 50 users, and concludes that the discrimination power of different feature is heavily user dependant i.e. while some features achieve 100% identification accuracy for some users, they perform poorly for others. On small smartphone 19% and on mini-tablet 42% of the users have unique FP compared to other features. We concluded that FP could be used alone to perfectly identify users for 100% accuracy.
fatcat:vdovp4xjojcgzetbd4i2dnuslu