Author, Institution: Mantas Bacevičius, Kaunas University of Technology
Science area, field of science: Natural Sciences, Informatics, N009
Research supervisor: Prof. Dr. Agnė Paulauskaitė (Kaunas University of Technology, Natural Sciences, Informatics, N009)
Dissertation Defence Board of Informatics Science Field:
Prof. Dr. Hab. Rimantas Barauskas (Kaunas University of Technology, Natural Sciences, Informatics, N009) – chairperson
Assoc. Prof. Dr. Mantas Lukoševičius (Kaunas University of Technology, Natural Sciences, Informatics, N009)
Prof. Dr. Gintaras Palubeckis (Kaunas University of Technology, Natural Sciences, Informatics, N009)
Prof. Dr. Simona Ramanauskaitė (Vilnius Gediminas Technical University, Natural Sciences, Informatics, N009)
Prof. Dr. Alexander Schlaefer (Hamburg University of Technology , Germany,
Natural Sciences, Informatics, N009)
Dissertation defence meeting will be at the Rectorate Hall of Kaunas University of Technology (K. Donelaičio 73-402, Kaunas)
The doctoral dissertation is available at the library of Kaunas University of Technology (Gedimino 50, Kaunas) and on the internet: M. Bacevičius el. dissertation.pdf
© M. Bacevičius, 2026 “The text of the thesis may not be copied, distributed, published, made public, including by making it publicly available on computer networks (Internet), reproduced in any form or by any means, including, but not limited to, electronic, mechanical or other means. Pursuant to Article 25(1) of the Law on Copyright and Related Rights of the Republic of Lithuania, a person with a disability who has difficulties in reading a document of a thesis published on the Internet, and insofar as this is justified by a particular disability, shall request that the document be made available in an alternative form by e-mail to doktorantura@ktu.lt.”
Annotation: The dissertation addresses one of the most pressing challenges in cybersecurity: how to ensure not only high accuracy in cyberattack detection, but also the explainability and reliability of the artificial intelligence (AI) solutions employed when working with high-dimensional and imbalanced network traffic data. Although deep learning methods demonstrate high classification accuracy, their practical application in critical infrastructure is limited by the “black-box” nature of their operation. Furthermore, the effectiveness of local explainable AI methods strongly depends on data characteristics and interdependencies among features, which often leads to unstable and unreliable explanations. The aim of this dissertation is to develop and experimentally investigate an explainable AI-based multiclass cyberattack detection methodology designed for the analysis of high-dimensional and class-imbalanced network traffic data, ensuring both high classification performance and the generation of stable and reliable explanations. The proposed solution integrates a hybrid Deep Autoencoder–Deep Neural Network (DAE-DNN) classifier together with a novel copula-based LIME (CoLIME) method. The developed DAE-DNN architecture achieves high classification performance (weighted F1-score of 0.9976) while ensuring 35–55% better throughput compared with other deep learning architectures. The novel CoLIME method enables the generation of explanations that more accurately reflect the classifier’s behavior. Experimental studies demonstrated that the proposed methodology improves explanation stability by 4.86–32.64% and fidelity by 13.26–188.64%.