Institutional Repository of Vilnius University
Biometriniai autentiškumo nustatymo metodai internetinėje mokymosi aplinkoje /
Abstract
dc:descriptionAlthough the COVID pandemic is over, not all educational institutions have returned to live learning and continue distance learning process. The advantages of this distance learning mode have led a number of educational institutions to use and improve their distance learning tools and ways of assessing students' knowledge. With the increasing use of such environments, security has become an increasingly important aspect and the identity authentication of the person or student being assessed during an exam has become a major challenge. The main objective of the thesis is to investigate keystroke-based authentication methods and evaluate their suitability for student authentication in online learning environments. The main tasks of the thesis are: to compare biometric authentication methods with keystroke-based authentication, to analyse keystroke-based authentication methods, and to select the most suitable models for students authenticating. The results of the experiments on the fixed data showed that the CNN with LSTM and 2 passwords in the authentication process achieved an EER accuracy of 3.91 \%. Based on the results of the experiments with dynamic data, the authentication process using GMM achieved an EER accuracy of 16.35 \%. Overfitting problem was not solved using artificial neural networks in free-text scenario. Additionally, authentication scheme was proposed for continuous student authentication.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2023
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Žarnauskas, Aretas,
Rights
dc:rights- Statement dc:rights
-
- info:eu-repo/semantics/openAccess
- Language dc:language
- lit
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://repository.vu.lt/VU:ELABAETD192843377&prefLang=en_US