Generalization-Aware Optimization-Based Feature Weight Learning for Hybrid Intrusion Detection Systems
International Journal of Engineering & Extended Technologies Research · 28.03.2026 · 08 (02)
DOI: 10.15662/ijeetr.2026.0802071 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Engineering & Extended Technologies Research, online erschienen am 28.03.2026. Verfasst von S. Mugambigai und 1 weiteren Autor:innen.
Abstract (Originalsprache)
Intrusion Detection Systems (IDS) are widely used to identify malicious activities in network traffic, yet many machine learning methods still rely on fixed feature selection that does not change during training and can lead to overfitting. This work presents a Generalization-Aware Optimization-Based Feature Weight Learning (OFWL) framework in which feature weights are updated during training using validation feedback instead of being fixed in advance. A key idea is that weight updates are accepted only when they improve validation loss, helping the model focus on generalization rather than simply fitting the training data. The learned feature representation is then used in a hybrid model that combines Logistic Regression (LR), Support Vector Machine (SVM), and Random Forest (RF). Feature stability is also examined through variance analysis, and statistical testing is used to check the consistency of the results. The method is evaluated on the KDDCup99 dataset and shows better performance than baseline models while keeping the computational cost low.
Zitieren
APA
Mugambigai, S., & Valarmathi, S. (2026). Generalization-Aware Optimization-Based Feature Weight Learning for Hybrid Intrusion Detection Systems. International Journal of Engineering & Extended Technologies Research, 08(02). https://doi.org/10.15662/ijeetr.2026.0802071
Vancouver
Mugambigai S, Valarmathi S. Generalization-Aware Optimization-Based Feature Weight Learning for Hybrid Intrusion Detection Systems. International Journal of Engineering & Extended Technologies Research. 2026;08(02). doi:10.15662/ijeetr.2026.0802071
BibTeX
@article{Mugambigai2026,
author = {Mugambigai, S. and Valarmathi, S.},
title = {Generalization-Aware Optimization-Based Feature Weight Learning for Hybrid Intrusion Detection Systems},
journal = {International Journal of Engineering & Extended Technologies Research},
year = {2026},
volume = {08},
number = {02},
doi = {10.15662/ijeetr.2026.0802071}
}
EUVIMED