EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Transformer Classifier in Adversarial Attack Assessment from Data Packets Using Interactive Modelling

Fatima Isiaka, Yunus Fatima Abdulsalam

International Journal of Computer Studies and Advancement in Current Research · 2026

DOI: 10.66419/ijcsacr.6668 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Computer Studies and Advancement in Current Research, 2026. Verfasst von Fatima Isiaka und 1 weiteren Autor:innen. Die Arbeit nennt 1 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This paper introduces an interactive modelling framework that leverages transformer classifiers to rigorously assess and mitigate adversarial attacks targeting data packet transmissions in network security. Traditional intrusion detection systems often struggle to identify sophisticated, stealthy perturbations embedded within high-velocity data streams. To address this vulnerability, our approach utilizes self-attention mechanisms inherent in transformer architectures to capture complex, long-range dependencies and subtle anomalies across sequential packet payloads. By integrating interactive modelling, human-in-the-loop validation is combined with automated threat simulation, allowing the classifier to dynamically adapt to evolving adversarial strategies in real time. Experimental evaluations demonstrate that the proposed transformer-based model significantly outperforms conventional machine learning baselines in detecting adversarial evasion techniques, offering a robust, interpretable, and resilient defense mechanism for next-generation communication networks.

Zitieren

APA

Isiaka, F., & Abdulsalam, Y. F. (2026). Transformer Classifier in Adversarial Attack Assessment from Data Packets Using Interactive Modelling. International Journal of Computer Studies and Advancement in Current Research. https://doi.org/10.66419/ijcsacr.6668

Vancouver

Isiaka F, Abdulsalam YF. Transformer Classifier in Adversarial Attack Assessment from Data Packets Using Interactive Modelling. International Journal of Computer Studies and Advancement in Current Research. 2026. doi:10.66419/ijcsacr.6668

BibTeX
@article{Isiaka2026,
  author = {Isiaka, Fatima and Abdulsalam, Yunus Fatima},
  title = {Transformer Classifier in Adversarial Attack Assessment from Data Packets Using Interactive Modelling},
  journal = {International Journal of Computer Studies and Advancement in Current Research},
  year = {2026},
  doi = {10.66419/ijcsacr.6668}
}
Kostenlos starten →