Zaher Hinbarji - Transfer Talk - 17th August 2015
Title: Behavioural Biometric Identification Based on Human-Computer Interactions
Supervisor: Dr. Cathal Gurrin & Dr. Rami Albatal
As we become increasingly dependent on information systems, personal identification and profiling systems have received an increasing interest, either for personalisation or security. Biometric profiling is one means of identification which can be achieved by analysing something the user is or does (e.g., a fingerprint, signature, face, voice). My focus is on behavioural biometrics, a subset of biometrics that is concerned with the patterns of conscious or unconscious behaviour of a person, involving their style, preference, skills, knowledge, motor-skills or strategy in any domain. In this work we explore the creation of user profiles to be applied in dynamic user identification based on the biometric patterns observed during normal human-computer interaction by continuously logging and tracking the corresponding low-level computer events. Specifically, we focus here on three components: mouse dynamics, keystrokes dynamics and GUI based user behaviour. Our focus is on exploring the application of such passive biometric sensing to identify the individual. In this talk I will describe my prior work on mouse-based behavioural biometrics, describe the LoggerMan application that passively gathers behavioural biometric data for my experiments, and introduce my research plan to integrate keystroke dynamics and GUI analytics to enhance the effectiveness of my profile building and dynamic user identification.