Generative AI Applications in Network Security, Monitoring, and Traffic Analysis
International Journal of Advanced Research in Science Communication and Technology · 30.06.2026 · 616
DOI: 10.48175/ijarsct-33287 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Advanced Research in Science Communication and Technology, online erschienen am 30.06.2026. Verfasst von Shallu Yadav. Die Arbeit nennt 40 Literaturangaben.
Abstract (Originalsprache)
The accelerating volume, velocity, and sophistication of modern network traffic have exposed the limitations of signature-based and conventional machine-learning defences, which struggle with previously unseen attacks, severe class imbalance, and the scarcity of labelled malicious samples. Generative Artificial Intelligence (Generative AI) has recently emerged as a transformative paradigm that addresses these gaps by learning the underlying distribution of benign and malicious traffic and producing realistic synthetic data, latent representations, and context-aware sequence models. This paper presents a comprehensive study of Generative AI applications across three tightly coupled domains: network security (intrusion and malware detection), network monitoring (anomaly and novelty detection), and traffic analysis (classification and behavioural profiling). We organise the principal generative families, Generative Adversarial Networks (GANs), Variational Autoencoders (VAEs), diffusion models, and Transformer-based large language models (LLMs), into a unified taxonomy and propose an end-to-end methodology in which generative augmentation is fused with discriminative detection. The framework is evaluated on three widely used benchmarks, NSL-KDD, CIC-IDS2017, and UNSW-NB15, against classical and deep-learning baselines. Experimental results show that generative augmentation raises detection accuracy by up to 7.9 percentage points over a Random-Forest baseline, lifts the area under the ROC curve to 0.988, and improves minority-class recall for rare attacks from 41.2% to over 83% when an appropriate synthetic-augmentation ratio is applied. We further analyse the trade-off between detection performance and computational cost, and discuss interpretability, adversarial robustness, and privacy-preserving data sharing. The findings confirm that Generative AI substantially strengthens data-driven network defence while highlighting open challenges in latency, stability, and trustworthy deployment.
Zitieren
APA
Yadav, S. (2026). Generative AI Applications in Network Security, Monitoring, and Traffic Analysis. International Journal of Advanced Research in Science Communication and Technology, 616. https://doi.org/10.48175/ijarsct-33287
Vancouver
Yadav S. Generative AI Applications in Network Security, Monitoring, and Traffic Analysis. International Journal of Advanced Research in Science Communication and Technology. 2026:616. doi:10.48175/ijarsct-33287
BibTeX
@article{Yadav2026,
author = {Yadav, Shallu},
title = {Generative AI Applications in Network Security, Monitoring, and Traffic Analysis},
journal = {International Journal of Advanced Research in Science Communication and Technology},
year = {2026},
pages = {616},
doi = {10.48175/ijarsct-33287}
}
EUVIMED