EUVIMEDEUVIMED entdecken
Crossref

An Optimised Transformer-Based Intrusion Detection Framework for IoT Networks Using Fisher Score Feature Selection and SMOTE Balancing Technique

Omeiza Abdulmumin Ganiyu, Olarinwaju Aliyu Hamzat, Femi Aminu Enesi

Faculty of Natural and Applied Sciences Journal of Computing and Applications · 31.03.2026 · 3 (1) · 130-136

DOI: 10.63561/jca.v3i1.1244 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Faculty of Natural and Applied Sciences Journal of Computing and Applications, online erschienen am 31.03.2026. Verfasst von Omeiza Abdulmumin Ganiyu und 2 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The proliferation of Internet of Things (IoT) devices has brought about significant advantages in areas such as smart homes and industrial automation. However, this widespread adoption has also rendered these systems more susceptible to cyberattacks, underscoring the critical need for robust intrusion detection systems. The proposed model presents an optimised approach to intrusion detection in IoT networks using a transformer-based architecture, renowned for its exceptional performance across diverse machine-learning tasks. This model was validated on the BoT-IoT Network Intrusion dataset, which encompasses a variety of IoT network traffic scenarios. The Synthetic Minority Oversampling Technique (SMOTE) was leveraged to address the imbalanced dataset. Furthermore, we improved efficiency by employing Fisher's score to select the most critical features. The Transformer-based model, leveraging the self-attention mechanism, was utilised to classify network traffic as normal or malicious. Our findings demonstrate the remarkable effectiveness of this approach, yielding an accuracy rate of 99%, precision of 99%, recall of 99%, F1 score of 99%, and an Area Under the Receiver Operating Characteristic Curve (AUC) of 99%. These scores confirm that our model can accurately detect and classify intrusions with minimal to no errors.

Zitieren

APA

Ganiyu, O. A., Hamzat, O. A., & Enesi, F. A. (2026). An Optimised Transformer-Based Intrusion Detection Framework for IoT Networks Using Fisher Score Feature Selection and SMOTE Balancing Technique. Faculty of Natural and Applied Sciences Journal of Computing and Applications, 3(1), 130-136. https://doi.org/10.63561/jca.v3i1.1244

Vancouver

Ganiyu OA, Hamzat OA, Enesi FA. An Optimised Transformer-Based Intrusion Detection Framework for IoT Networks Using Fisher Score Feature Selection and SMOTE Balancing Technique. Faculty of Natural and Applied Sciences Journal of Computing and Applications. 2026;3(1):130-136. doi:10.63561/jca.v3i1.1244

BibTeX
@article{Ganiyu2026,
  author = {Ganiyu, Omeiza Abdulmumin and Hamzat, Olarinwaju Aliyu and Enesi, Femi Aminu},
  title = {An Optimised Transformer-Based Intrusion Detection Framework for IoT Networks Using Fisher Score Feature Selection and SMOTE Balancing Technique},
  journal = {Faculty of Natural and Applied Sciences Journal of Computing and Applications},
  year = {2026},
  volume = {3},
  number = {1},
  pages = {130--136},
  doi = {10.63561/jca.v3i1.1244}
}
Kostenlos starten →