A novel security scheme for behavioral authentication systems based on keystroke dynamics

Asma Salem, Mohammad S. Obaidat

2019

Closed access Metadata (CSL-JSON)
DOI 10.1002/spy2.64
Data confidence 70 medium
How this score is calculated

It is built from checkable properties of the record itself: whether the article has a DOI, how many independent sources confirmed its data, whether the required fields are filled in, and how recent the latest confirmation is. Disagreement between sources and a missing identifier lower it. A value of 80 or above means the record was published without further checking; 60 to 79 means it is published at reduced confidence, with a manual check advisable. A record verified by an editor by hand is never pushed below 90 by a recalculation. This scores our data about the article, not the article itself.

Abstract

This article presents a comprehensive analysis behavioral authentication systems based on keystroke dynamics using android mobile devices. Behavioral security authentication is an efficient biometric‐based security system that can be used to authenticate users. It is exploited to strengthen password authentication efficiently and inexpensively because no extra hardware is required in most of these schemes. Keystroke dynamic rhythm uses combinations of timing features and nontiming features that are extracted and processed from several devices such as classical keyboard and built‐in software keyboard in touch screen and smart phone devices. This work presents a comprehensive analysis of using biometric behavioral authentication system‐based on keystroke dynamics. Neural network classifiers are used in this work. The performance results show that keystroke dynamics provides good level in performance measures as a second authentication factor. The distinguishable role for nontiming features in addition to the timing features is demonstrated. These features have a significant role in improving the performance measures. The proposed model achieves low error rate of 0.3% for false acceptance rate, a false rejection of 1.5%, and an equal error rate of 0.9%. These are considered excellent enhancements when compared with previous reported results. Suggestions for future directions, and challenges for using behavioral‐based authentication systems are also highlighted.