EUVIMEDEUVIMED entdecken
Crossref

A Hybrid GNN-Transformer Framework for Enhanced Detection of Distributed Denial-of-Service (DDoS) Attacks in Network Traffic

Mustafa Al-Mayyahi

Dijlah Journal of Engineering Sciences ISSN: 3078-9664, e-ISSN: 3078-9656 · 2026

DOI: 10.65204/djes.v3i2.583 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Dijlah Journal of Engineering Sciences ISSN: 3078-9664, e-ISSN: 3078-9656, 2026. Verfasst von Mustafa Al-Mayyahi.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

One of the most important threats of security in new networks refers to Distributed Denial-of-Service (DDoS) attacks, as their wide data amount and complexity make their diagnosis difficult. Present research offers multiple architecture given the Graph Neural Networks (GNNs) and Transformer models for DDoS attack diagnosis. Firstly, data of network is schemed as a graph, nodes and edges’ attributes are extracted applying GNN. After that, model of Transformer is developed for analyzing temporal dependencies as well as extracting models associated with shared attacks. Such architecture leverages GNN to learn complicated structures of graph and Transformer to analyze long-term relations in temporal data. Outcomes of experiments on UNSW-NB15 dataset show that presented architecture excels in diagnosing DDoS attack decreasing attributes’ amount when developing accuracy of diagnosis. Model obtains an F1 score of 98.3%, accuracy of 98.7%, a recall of 97.9% illustrating high ability for network intrusion detection systems (IDS) usage.

Zitieren

APA

Al-Mayyahi, M. (2026). A Hybrid GNN-Transformer Framework for Enhanced Detection of Distributed Denial-of-Service (DDoS) Attacks in Network Traffic. Dijlah Journal of Engineering Sciences ISSN: 3078-9664, e-ISSN: 3078-9656. https://doi.org/10.65204/djes.v3i2.583

Vancouver

Al-Mayyahi M. A Hybrid GNN-Transformer Framework for Enhanced Detection of Distributed Denial-of-Service (DDoS) Attacks in Network Traffic. Dijlah Journal of Engineering Sciences ISSN: 3078-9664, e-ISSN: 3078-9656. 2026. doi:10.65204/djes.v3i2.583

BibTeX
@article{AlMayyahi2026,
  author = {Al-Mayyahi, Mustafa},
  title = {A Hybrid GNN-Transformer Framework for Enhanced Detection of Distributed Denial-of-Service (DDoS) Attacks in Network Traffic},
  journal = {Dijlah Journal of Engineering Sciences ISSN: 3078-9664, e-ISSN: 3078-9656},
  year = {2026},
  doi = {10.65204/djes.v3i2.583}
}
Kostenlos starten →