SHAP TEXNOLOGIYASI ASOSIDA DDOS HUJUMLARINI ANIQLASH MODELINI OPTIMALLASHTIRISH
DOI:
https://doi.org/10.65164/5tggpz92Ключевые слова:
DDoS hujumlari, SHAP, feature selection, Random Forest, tarmoq xavfsizligi, model optimallashtirish, explainable AI.Аннотация
Ushbu maqolada SHAP (SHapley Additive exPlanations) texnologiyasi
yordamida DDoS hujumlarini aniqlash modelini optimallashtirish usuli taqdim etiladi. Uchta xalqaro
benchmark dataset (CIC-DDoS2019, UNSW-NB15, Network Intrusion) ustida o'tkazilgan tajribalar
shuni ko'rsatdiki, SHAP asosida tanlangan 10 ta belgi bilan o'qitilgan model to'liq belgilar to'plamiga
nisbatan CIC-DDoS2019 da +0.82%, UNSW-NB15 da +2.04% yuqori aniqlik ko'rsatdi. Belgilar soni
esa 74–87% ga qisqardi. Natijalar SHAP texnologiyasining DDoS aniqlash tizimlarida samarali
feature selection usuli sifatida qo'llanilishi mumkinligini isbotlaydi.
Библиографические ссылки
[1]. Waqas M. va boshq., "A Systematic Review of DDoS Attack Detection Techniques". – IEEE
Access, 2022, 10, 45891–45924 b.
[2]. Lundberg S.M., Lee S.I., "A Unified Approach to Interpreting Model Predictions". – NeurIPS,
2017, 4765–4774 b.
[3]. Sharafaldin I., Lashkari A.H., Ghorbani A.A., "Toward Generating a New Intrusion Detection
Dataset". – ICISSP, 2018, 108–116 b.
[4]. Moustafa N., Slay J., "UNSW-NB15: A Comprehensive Data Set for Network Intrusion
Detection Systems". – MilCIS, 2015 b.
[5]. Anand V. va boshq., "Network Intrusion Detection Using SHAP-based Explainable AI". –
IEEE ICCCNT, 2021, 1–6 b.
[6]. Farnaaz N., Jabbar M.A., "Random Forest Modeling for Network Intrusion Detection System".
– Procedia Computer Science, 2016, 89, 213–217 b.
