A Systematic Review of Hybrid Deep Learning and Optimisation Techniques to Address Class Imbalance and Malware Detection in IoT Networks
Pertanika Journal of Science and Technology · 30.10.2026 · 34 (5)
DOI: 10.47836/pjst.34.5.15 ↗Kurz gefasst
Systematische Übersicht in der Zeitschrift Pertanika Journal of Science and Technology, online erschienen am 30.10.2026. Verfasst von Ernest Ezema und 1 weiteren Autor:innen.
Abstract (Originalsprache)
The Internet of Things (IoT) is a rapidly expanding technology, resulting in an increasingly complex cybersecurity environment for malware detection. Network traffic classification (NTC) remains a challenge. Traditional detection techniques are ineffective at detecting encrypted traffic when attackers change their behaviour and when intrusion datasets exhibit class imbalance. The systematic literature review, conducted in accordance with the PRISMA 2020 framework, presents the results of three primary peer-reviewed studies (2019-2025). The corpus was constructed through a systematic search of IEEE Xplore, ScienceDirect, ACM Digital Library, Scopus, and SpringerLink, using strict inclusion and exclusion criteria. Six thematic dimensions are explored: IoT threat landscape, malware detection strategies, hybrid deep learning architectures, optimisation techniques for class imbalance, feature engineering, and evaluation practices. Key findings indicate that integrating CNNs and LSTMs, PSO-based classifiers, attention mechanisms, and federated learning models is a cutting-edge area in IoT security research. There are significant challenges in adaptive sampling, zero-day generalisation, encrypted traffic classification, scalability for resource-constrained environments and real-world testbed validation. To address these needs, this review presents a conceptual framework that combines GA/PSO-guided adaptive sampling with a deep learning pipeline evaluated on a simulated IoT testbed. The study concludes with a structured research roadmap for future work on explainability, real-time adaptability, and lightweight security mechanisms.
Zitieren
APA
Ezema, E., & Juremi, J. (2026). A Systematic Review of Hybrid Deep Learning and Optimisation Techniques to Address Class Imbalance and Malware Detection in IoT Networks. Pertanika Journal of Science and Technology, 34(5). https://doi.org/10.47836/pjst.34.5.15
Vancouver
Ezema E, Juremi J. A Systematic Review of Hybrid Deep Learning and Optimisation Techniques to Address Class Imbalance and Malware Detection in IoT Networks. Pertanika Journal of Science and Technology. 2026;34(5). doi:10.47836/pjst.34.5.15
BibTeX
@article{Ezema2026,
author = {Ezema, Ernest and Juremi, Julia},
title = {A Systematic Review of Hybrid Deep Learning and Optimisation Techniques to Address Class Imbalance and Malware Detection in IoT Networks},
journal = {Pertanika Journal of Science and Technology},
year = {2026},
volume = {34},
number = {5},
doi = {10.47836/pjst.34.5.15}
}
EUVIMED