Adaptive Intrusion Detection System Using LightGBM and Explainable Feature Selection for Network Security
International Journal of Innovative Science and Research Technology · 19.06.2026 · 4485
DOI: 10.38124/ijisrt/26may2139 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Innovative Science and Research Technology, online erschienen am 19.06.2026. Verfasst von Bibhuti Bhusan Parida und 4 weiteren Autor:innen. Die Arbeit nennt 17 Literaturangaben.
Abstract (Originalsprache)
The rapid growth of network-based applications has significantly increased the risk of cyber-attacks, making accurate and efficient intrusion detection systems essential for modern network security. Traditional intrusion detection approaches often suffer from high computational complexity, limited scalability, and reduced detection performance when dealing with large-scale network traffic. This paper presents an adaptive intrusion detection framework based on the Light Gradient Boosting Machine (LightGBM) algorithm for efficient classification of network intrusions. The proposed framework employs data preprocessing, feature engineering, and feature importance analysis to identify the most relevant network traffic attributes from the UNSW-NB15 dataset. LightGBM is utilized as the primary classification engine due to its fast-training capability, low memory consumption, and superior predictive performance. To improve interpretability and facilitate security analysis, feature importance scores are incorporated to explain attack detection decisions. The performance of the proposed model is evaluated using accuracy, precision, recall, F1-score, and receiver operating characteristic metrics. Experimental results demonstrate that the proposed approach achieves high detection accuracy while maintaining low computational overhead, making it suitable for real-time intrusion detection applications. Comparative analysis with conventional machine learning classifiers indicates the effectiveness and robustness of the proposed LightGBM-based intrusion detection framework for modern cybersecurity environments.
Zitieren
APA
Parida, B. B., Besra, B., Hembram, B., Singh, B., & Munda, B. S. (2026). Adaptive Intrusion Detection System Using LightGBM and Explainable Feature Selection for Network Security. International Journal of Innovative Science and Research Technology, 4485. https://doi.org/10.38124/ijisrt/26may2139
Vancouver
Parida BB, Besra B, Hembram B, Singh B, Munda BS. Adaptive Intrusion Detection System Using LightGBM and Explainable Feature Selection for Network Security. International Journal of Innovative Science and Research Technology. 2026:4485. doi:10.38124/ijisrt/26may2139
BibTeX
@article{Parida2026,
author = {Parida, Bibhuti Bhusan and Besra, Basta and Hembram, Basta and Singh, Binod and Munda, Binu Singh},
title = {Adaptive Intrusion Detection System Using LightGBM and Explainable Feature Selection for Network Security},
journal = {International Journal of Innovative Science and Research Technology},
year = {2026},
pages = {4485},
doi = {10.38124/ijisrt/26may2139}
}
EUVIMED