Combining Mouse and Eye Movement Biometrics for User Authentication

Lü, Hongwei; Rose, J. T. Anita; Liu, Yudong; Awad, Ahmed; Hou, Leon

doi:10.1007/978-3-319-48947-6_5

Cited by 7 publications

(5 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The performance of the proposed approach was evaluated on a dataset collected from forty users. The results clarified that coupling eye tracking with the mouse dynamics are applicable for authentication [9].…”

Section: Issn: 0067-2904mentioning

confidence: 86%

User Authentication via Mouse Dynamics

Salman

Hameed

2018

ijs

View full text Add to dashboard Cite

Nowadays, the development of internet communication and the significant increase of using computer lead in turn to increasing unauthorized access. The behavioral biometric namely mouse dynamics is one means of achieving biometric authentication to safeguard against unauthorized access. In this paper, user authentication models via mouse dynamics to distinguish users into genuine and imposter are proposed. The performance of the proposed models is evaluated using a public dataset consists of 48 users as an evaluation data, where the Accuracy

show abstract

Section: Issn: 0067-2904mentioning

confidence: 86%

User Authentication via Mouse Dynamics

Salman

Hameed

2018

ijs

View full text Add to dashboard Cite

show abstract

“…Several works have investigated closely the utility of mouse cursor data in web search [27][28][29] and web page usability evaluation [30][31][32], two of the most prominent use cases of this technology. Mouse biometrics is another active research area that has recently shown how to identify an individual by analyzing their mouse movements in controlled settings [33,34]. This is the research line we focus on in this work.…”

Section: Biometrics Prediction From Movement Datamentioning

confidence: 99%

What can a swiped word tell us more? Demographic and behavioral correlates from shape-writing text entry

Lemarquis

Yilma

Leiva

2023

Neural Comput & Applic

View full text Add to dashboard Cite

Shape-writing (aka gesture typing or swiping) is a word-based text entry method for touchscreen keyboards. It works by landing the finger on (or close to) the first character of the desired word and then sliding over all the other character keys without lifting the finger until the last word character is reached. This generates a trajectory of swiped characters on the keyboard layout which can be translated to a meaningful word by a statistical decoder. We hypothesize that swiping carries rich information about the user, such as demographic (e.g. age or gender) and behavioral (e.g. swiping familiarity or input finger) information. To test our hypothesis, we trained several sequence classifiers using different recurrent neural network architectures to predict demographic and behavioral correlates of users from swipe trajectories. We show that our sequence classifiers are always performing better than a random classifier, therefore we conclude that cognitive and motor control mechanisms are embodied and reflected in swipe trajectories, validating thus our research hypothesis. Taken together, our results have implications for user privacy. Currently swiping is supported by all mobile vendors and has millions of users, so people may be inadvertently profiled at an unprecedented granularity. Future work should consider new ways of addressing these issues without impacting the user's swiping experience.

show abstract

“…Alternative preprocessing methods or simply more data could lead to better accuracies. For example, many classification problems can also be approached using regression techniques, and mouse dynamics is not an exception [40]. Further exploring the varied user frequency in the top-10 accuracies for each model could also reveal hidden aspects of the data that can be employed to further improve results.…”

Section: Limitationsmentioning

confidence: 99%

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

Siddiqui

Dave

Vanamala

et al. 2022

MAKE

View full text Add to dashboard Cite

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 user’s 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 three 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 (1D-CNN) 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 (ANN) which reaches an astounding peak accuracy of 92.48%, the highest accuracy we have seen for any classifier on this dataset.

show abstract

Combining Mouse and Eye Movement Biometrics for User Authentication

Cited by 7 publications

References 12 publications

User Authentication via Mouse Dynamics

User Authentication via Mouse Dynamics

What can a swiped word tell us more? Demographic and behavioral correlates from shape-writing text entry

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

Contact Info

Product

Resources

About