TKCA: a timely keystroke-based continuous user authentication with short keystroke sequence in uncontrolled settings

Abstract

Keystroke-based behavioral biometrics have been proven effective for continuous user authentication. Current state-of-the-art algorithms have achieved outstanding results in long text or short text collected by doing some tasks. It remains a considerable challenge to authenticate users continuously and accurately with short keystroke inputs collected in uncontrolled settings. In this work, we propose a Timely Keystroke-based method for Continuous user Authentication, named TKCA. It integrates the key name and two kinds of timing features through an embedding mechanism. And it captures the relationship between context keystrokes by the Bidirectional Long Short-Term Memory (Bi-LSTM) network. We conduct a series of experiments to validate it on a public dataset - the Clarkson II dataset collected in a completely uncontrolled and natural setting. Experiment results show that the proposed TKCA achieves state-of-the-art performance with 8.28% of EER when using only 30 keystrokes and 2.78% of EER when using 190 keystrokes.

Internet Users' Willingness to Disclose Biometric Data for Continuous Online Account Protection: An Empirical Investigation

“…Keystroke Dynamics [100] The way you use a keyboard (e.g., how long you hold down a certain key). Mouse Dynamics [101] The way you use a mouse (e.g., how long you hold down a mouse button or how fast you move the mouse pointer).…”

Keystroke Dynamics based Recognition Systems using Deep Learning: A Survey

“…A detailed comparative analysis was presented based on two points: the biometrics used and deep learning methods applied, to justify accomplishments of research. Current proposed work claimed to obtain 8.28% EER with 30 keystrokes and 2.78% EER with 190 keystrokes [51]. A continuous authentication based fraudulent detection system (KOLLECTOR) was developed in a research work by applying Deep learning concept: GRU-BRNN (Gated Recurrent Unit-Bidirectional Recurrent Neural Network), for mobile phones.…”

“…Total 39 subjects participated for keystroke sample collection in two sessions (11 months duration), and on average 21533 characters were typed by each subject. Used in research: [59], [61], [50], [51]. Keystroke Dynamics on Android platform [63]: Fixed and short passcode ".tie5Ronal" was typed by 42 users, where each user provides 51 keystroke timing samples.…”