Skip to Main Content
The mouse dynamics biometric is a behavioral biometric technology that extracts and analyzes the movement characteristics of the mouse input device when a computer user interacts with a graphical user interface for identification purposes. Most of the existing studies on mouse dynamics analysis have targeted primarily continuous authentication or user reauthentication for which promising results have been achieved. Static authentication (at login time) using mouse dynamics, however, appears to face some challenges due to the limited amount of data that can reasonably be captured during such a process. In this paper, we present a new mouse dynamics analysis framework that uses mouse gesture dynamics for static authentication. The captured gestures are analyzed using a learning vector quantization neural network classifier. We conduct an experimental evaluation of our framework with 39 users, in which we achieve a false acceptance ratio of 5.26% and a false rejection ratio of 4.59% when four gestures were combined, with a test session length of 26.9 s. This is an improvement both in the accuracy and validation sample, compared to the existing mouse dynamics approaches that could be considered adequate for static authentication. Furthermore, to our knowledge, our work is the first to present a relatively accurate static authentication scheme based on mouse gesture dynamics.