DEEP LEARNING-BASED MALICIOUS TRAFFIC ANALYSIS: A COMPREHENSIVE SURVEY
Journal of Computer Science and Electrical Engineering · 22.05.2026 · 8 (4) · 1-16
DOI: 10.61784/jcsee3136 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Computer Science and Electrical Engineering, online erschienen am 22.05.2026. Verfasst von WenCai He und 3 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Malicious traffic analysis has become increasingly challenging due to encryption-by-default communication, evolving attack behaviors, and distribution shifts across heterogeneous environments. At present, many studies still do not share consistent pipeline designs, traffic representations, benchmark datasets, or assessment methods. Based on this problem, this paper systematically examines deep-learning-based malicious-traffic-analysis technologies. Existing works are organized into four main parts: data acquisition and preprocessing, traffic representation, learning models, and performance evaluation methods. In particular, we compare representative methods, including traditional machine learning, deep learning, and hybrid forms, in terms of detection accuracy, computational cost, cross-scenario generalization ability, and suitable metrics. We also identify key problems in practical applications, such as encrypted traffic monitoring capability, biased datasets, and adversarial attacks, and summarize related research paths. This paper introduces the research area and provides a reference for reproducible literature-based evaluation design.
Zitieren
APA
He, W., Peng, Z., Zhang, M., & Zhu, H. (2026). DEEP LEARNING-BASED MALICIOUS TRAFFIC ANALYSIS: A COMPREHENSIVE SURVEY. Journal of Computer Science and Electrical Engineering, 8(4), 1-16. https://doi.org/10.61784/jcsee3136
Vancouver
He W, Peng Z, Zhang M, Zhu H. DEEP LEARNING-BASED MALICIOUS TRAFFIC ANALYSIS: A COMPREHENSIVE SURVEY. Journal of Computer Science and Electrical Engineering. 2026;8(4):1-16. doi:10.61784/jcsee3136
BibTeX
@article{He2026,
author = {He, WenCai and Peng, ZhiJie and Zhang, MingShen and Zhu, He},
title = {DEEP LEARNING-BASED MALICIOUS TRAFFIC ANALYSIS: A COMPREHENSIVE SURVEY},
journal = {Journal of Computer Science and Electrical Engineering},
year = {2026},
volume = {8},
number = {4},
pages = {1--16},
doi = {10.61784/jcsee3136}
}
EUVIMED