Securing Sensitive Information in Smart Mobile Devices through Difficult-to-Mimic and Single-Time Usage Analytics10/04/2017 Objective: Inspiration This paper presents a new very lightweight single-time machine learning based user identification technique, which is near-to-impossible to breach by intruders. The performance evaluation over 33 participants confirms as good as only 5% FAR and only 6% FRR. Further implementation as a mobile application demonstrates only 0% FAR in real-time usage. Speaker(s)
|