Optimizing Intrusion Detection Systems through Benchmarking of Ensemble Classifiers on Diverse Network Attacks
STAP Journal of Security Risk Management · 2025
DOI: 10.63180/jsrm.thestap.2025.1.4 ↗Kurz gefasst
Fachartikel in der Zeitschrift STAP Journal of Security Risk Management, 2025. Verfasst von Dena Abu Laila und 3 weiteren Autor:innen. Die Arbeit wurde bisher 5-mal zitiert (Crossref).
Abstract (Originalsprache)
The escalating sophistication of cyber threats requires transparent and reproducible benchmarks for intelligent security paradigms. This study presents a comprehensive benchmark analysis of a machine learning pipeline for network intrusion detection, addressing critical deployment oriented challenges such as class imbalance, feature optimization, and cross-environment generalization. Trained rigorously on the NF-CSE-CIC-IDS2018-v2 dataset and validated on the distinct UNSWNB15 dataset, this work tackles the complexities of identifying diverse network threats through the systematic integration of data preprocessing, advanced class-imbalance handling with SMOTE, and an embedded feature selection methodology. A comparative evaluation is conducted between state-of-the-art ensemble models (Random Forest and XGBoost), recent deep learning approaches, and a logistic regression baseline, examining predictive accuracy, computational trade-offs, and per-class performance across stealthy and volumetric attack types. The optimized Random Forest model achieves 99.95% accuracy and a 0.9837 F1-score on the primary dataset, while demonstrating strong generalization performance with a 94.8% F1-score on cross-validation, supported by thorough overfitting analysis and model validation procedures.
Zitieren
APA
Laila, D. A., Aljawarneh, M., Al-Na’amneh, Q., & Sulaiman, R. B. (2025). Optimizing Intrusion Detection Systems through Benchmarking of Ensemble Classifiers on Diverse Network Attacks. STAP Journal of Security Risk Management. https://doi.org/10.63180/jsrm.thestap.2025.1.4
Vancouver
Laila DA, Aljawarneh M, Al-Na’amneh Q, Sulaiman RB. Optimizing Intrusion Detection Systems through Benchmarking of Ensemble Classifiers on Diverse Network Attacks. STAP Journal of Security Risk Management. 2025. doi:10.63180/jsrm.thestap.2025.1.4
BibTeX
@article{Laila2025,
author = {Laila, Dena Abu and Aljawarneh, Mahmoud and Al-Na’amneh, Qais and Sulaiman, Rejwan Bin},
title = {Optimizing Intrusion Detection Systems through Benchmarking of Ensemble Classifiers on Diverse Network Attacks},
journal = {STAP Journal of Security Risk Management},
year = {2025},
doi = {10.63180/jsrm.thestap.2025.1.4}
}
EUVIMED