Threshold Selection and Attack Type Generalization in XGBoost and Denoising Autoencoder Intrusion Detection
Information · 16.11.2026 · 17 (11) · 1155
DOI: 10.3390/info17111155 ↗Kurz gefasst
Fachartikel in der Zeitschrift Information, online erschienen am 16.11.2026. Verfasst von Abildaeva Zhansaya und 7 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Hybrid intrusion detectors must divide a false-positive budget between supervised and anomaly scores. We study allocation stability, changed alerts, and attack-type transfer using extreme gradient boosting (XGBoost) and a benign-trained denoising autoencoder (DAE). Experiments include cross-day evaluation, fixed-model resampling, type and grouped-family holdouts, matched baselines, and corrected CICIDS2017 and UNSW-NB15 replications. At the 1% calibration target, exploratory Friday recall is 67.46% for DAE alone versus 47.47% for selected fusion, with observed false-positive rates of 7.29% and 5.69%. With re-estimated thresholds, original CICIDS2017 allocations change in 25.0–54.0% of development resamples at the 0.1% target and 65.0–78.5% at 1%; mean retrospective Friday prediction disagreement at 1% is only 0.81–1.04%. Selected fusion gains 5.86 percentage points of macro recall over XGBoost in six individual-type holdouts at 1%. With all four denial-of-service (DoS) types excluded together, however, 1% within-family recall is 19.26% for selected fusion, 14.25% for XGBoost, and 54.78% for DAE; both-Patator exclusion gives 5.46% for selected fusion versus 7.66% for XGBoost. Ordinary-autoencoder fusion performs comparably to DAE fusion in the individual-type protocol, and longer denoising training does not consistently improve detection. On corrected CICIDS2017, selected fusion reaches 66.18% macro recall versus 70.71% for XGBoost. UNSW-NB15 holdouts disable the anomaly branch throughout, with a 4.38% mean test false-positive rate at the 1% target; a separate all-known audit finds rare switching. The contribution separates allocation variability, changed decisions, and budget transfer. Neither stable selection nor gains on individual attack types establish reliable unseen-family detection.
Zitieren
APA
Zhansaya, A., Konyrbayev, N., Dauitbayeva, A., Elmira, T., Almenova, A., Kazhymurat, G., Karim, N., & Cho, Y. I. (2026). Threshold Selection and Attack Type Generalization in XGBoost and Denoising Autoencoder Intrusion Detection. Information, 17(11), 1155. https://doi.org/10.3390/info17111155
Vancouver
Zhansaya A, Konyrbayev N, Dauitbayeva A, Elmira T, Almenova A, Kazhymurat G, et al. Threshold Selection and Attack Type Generalization in XGBoost and Denoising Autoencoder Intrusion Detection. Information. 2026;17(11):1155. doi:10.3390/info17111155
BibTeX
@article{Zhansaya2026,
author = {Zhansaya, Abildaeva and Konyrbayev, Nurbek and Dauitbayeva, Aigul and Elmira, Tulegenova and Almenova, Akmaral and Kazhymurat, Galymzhan and Karim, Nazerke and Cho, Young Im},
title = {Threshold Selection and Attack Type Generalization in XGBoost and Denoising Autoencoder Intrusion Detection},
journal = {Information},
year = {2026},
volume = {17},
number = {11},
pages = {1155},
doi = {10.3390/info17111155}
}
EUVIMED