Feature-Driven Fusion of Browser Vulnerabilities and Session Indicators for Intelligent Intrusion Detection
Nature Journal of Emerging Sciences Technologies and Innovations · 2026
DOI: 10.65752/w0z8ye50 ↗Kurz gefasst
Fachartikel in der Zeitschrift Nature Journal of Emerging Sciences Technologies and Innovations, 2026. Verfasst von Mary Ofuru Kama und 5 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref). Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
The increasing sophistication of browser-based cyberattacks, such as session hijacking and drive-by downloads, highlights critical limitations in intrusion detection systems (IDS) that rely solely on detection of intruders through network or host-level signals without consideration to behavioral patterns. This work presents a feature-driven fusion framework that integrates behavioral and device-level indicators to enhance the detection of advanced browser intrusions. Behavioral features quantify user-session dynamics, while device-level metrics capture vulnerable exposures including browser versions and plugin security states. The framework employs ensemble learning and feature attribution analyzing to identify malicious activity with high accuracy. Evaluation on hybrid data sets combining HTPP logs or unknown browser logs and device telemetry shows that Random Forest (RF) and Decision Tree (DT) models outperformed other models with precision value of 0.906, recall of 0.886 and F1-score of 0.883. In contrast, Logistic regression (LR) recorded the accuracy values of precision to be 0.728, recall of 0.729, and F1-score 0.727, demonstrating the superiority of ensemble-based approaches for complect intrusion detection. The research provided a transparent, explainable pipeline for cross-layer threat detection in modern web environments, offering practical insights for improving IDS performance and resilience against sophisticated browser attacks. These results highlight the value of integrating behavioral and device-level features in security frameworks to support accurate and interpretable threat detection.
Zitieren
APA
Kama, M. O., Ugah, J., Oyiri, D., Eziaha, I., Ikegwu, A. C., & Anyanwu, C. M. (2026). Feature-Driven Fusion of Browser Vulnerabilities and Session Indicators for Intelligent Intrusion Detection. Nature Journal of Emerging Sciences Technologies and Innovations. https://doi.org/10.65752/w0z8ye50
Vancouver
Kama MO, Ugah J, Oyiri D, Eziaha I, Ikegwu AC, Anyanwu CM. Feature-Driven Fusion of Browser Vulnerabilities and Session Indicators for Intelligent Intrusion Detection. Nature Journal of Emerging Sciences Technologies and Innovations. 2026. doi:10.65752/w0z8ye50
BibTeX
@article{Kama2026,
author = {Kama, Mary Ofuru and Ugah, John and Oyiri, Dike and Eziaha, Ibegbulem and Ikegwu, Anayo Chukwu and Anyanwu, Chinatu Michael},
title = {Feature-Driven Fusion of Browser Vulnerabilities and Session Indicators for Intelligent Intrusion Detection},
journal = {Nature Journal of Emerging Sciences Technologies and Innovations},
year = {2026},
doi = {10.65752/w0z8ye50}
}
EUVIMED