Machine and Deep Learning Applications to Mouse Dynamics for Continuous User Authentication

TL;DR

This paper evaluates machine learning and deep learning models on mouse dynamics data for continuous user authentication, achieving high accuracy in identifying users and verifying identities.

Contribution

It introduces a comprehensive evaluation of multiple algorithms on a new dataset for mouse dynamics-based authentication, highlighting the effectiveness of CNNs and neural networks.

Findings

CNN achieved 85.73% accuracy in binary user authentication.

ANN reached 92.48% accuracy in multi-class user classification.

The study demonstrates mouse dynamics as a promising biometric for continuous authentication.

Abstract

Static authentication methods, like passwords, grow increasingly weak with advancements in technology and attack strategies. Continuous authentication has been proposed as a solution, in which users who have gained access to an account are still monitored in order to continuously verify that the user is not an imposter who had access to the user credentials. Mouse dynamics is the behavior of a users mouse movements and is a biometric that has shown great promise for continuous authentication schemes. This article builds upon our previous published work by evaluating our dataset of 40 users using three machine learning and deep learning algorithms. Two evaluation scenarios are considered: binary classifiers are used for user authentication, with the top performer being a 1-dimensional convolutional neural network with a peak average test accuracy of 85.73% across the top 10 users. Multi class classification is also examined using an artificial neural network which reaches an astounding peak accuracy of 92.48% the highest accuracy we have seen for any classifier on this dataset.