KIBERHUJUMLARNI ANIQLASH UCHUN MASHINAVIY O‘RGANISH MODELLARI YARATISH
Keywords:
Kalit so‘zlar: kiberhujumlar, IDS, mashinaviy o‘rganish, XGBoost, Random Forest, NSL-KDD, CICIDS-2017, SMOTE, Bayesian Optimization, real vaqt aniqlash, tarmoq xavfsizligi.Abstract
Annotatsiya: Ushbu maqolada kiberhujumlarni aniqlash uchun mashinaviy o‘rganish modellari yaratish masalasi ilmiy-amaliy nuqtai nazardan tahlil qilinadi. Raqamli iqtisodiyot sharoitida kiberhujumlar soni va murakkabligi ortib borayotgani an’anaviy imzoga asoslangan aniqlash usullarining imkoniyatlarini cheklab qo‘ymoqda. Tadqiqot doirasida NSL-KDD va CICIDS-2017 benchmark ma’lumotlar to‘plamlari asosida Logistik regressiya, Random Forest, XGBoost va ko‘p qavatli neyron tarmoq modellari qurildi hamda 5 qatlamli stratifikatsiyalangan kross-validatsiya protokolida baholandi. Ma’lumotlarni tayyorlash bosqichida Label Encoding, dispersiya filtri, korrelyatsiya filtri, Mutual Information asosida xususiyatlar tanlash, StandardScaler normalizatsiya va SMOTE algoritmi qo‘llanildi. Bayesian Optimization yordamida XGBoost giperparametrlari optimallashtirildi va model NSL-KDD to‘plamida 99,21% aniqlik, 99,11% F1-score hamda 1,8 ms inference vaqtiga erishdi. Quantum Shield real vaqtli monitoring tizimi Windows muhitida 72 soatlik sinovdan o‘tkazilib, 87,8% haqiqiy musbat darajasi va atigi 0,5% noto‘g‘ri musbat ko‘rsatkichiga erishdi. Olingan natijalar mashinaviy o‘rganish asosidagi yondashuvlarning kiberhujumlarni aniqlashdagi samaradorligini tasdiqlaydi.
References
1. UZ-CERT. O‘zbekiston Respublikasi Kompyuter Hodisalariga Javob Berish Guruhi yillik hisobotlari 2021–2023. Toshkent, 2024.
2. Buczak, A.L., Guven, E. A Survey of Data Mining and Machine Learning Methods for Cyber Security Intrusion Detection. IEEE Communications Surveys & Tutorials, 2016, Vol. 18, No. 2, pp. 1153–1176.
3. Sommer, R., Paxson, V. Outside the Closed World: On Using Machine Learning for Network Intrusion Detection. IEEE Symposium on Security and Privacy, 2010, pp. 305–316.
4. NIST Special Publication 800-94 Rev.1. Guide to Intrusion Detection and Prevention Systems (IDPS). National Institute of Standards and Technology, 2023.
5. Khaitbayev A. P. (2025). Deep learnıng vs tradıtıonal supervısed models ın ınformatıon securıty monıtorıng– Scientific Journal of Construction and education. – 109 b.
6. Anderson, B. et al. Machine Learning for Encrypted Malware Traffic Classification. ACM SIGKDD, 2020, pp. 1–10.
7. Tavallaee, M., Bagheri, E., Lu, W., Ghorbani, A. A Detailed Analysis of the KDD CUP 99 Data Set. IEEE CISDA, 2009, pp. 1–6.
8. Sharafaldin, I., Lashkari, A.H., Ghorbani, A.A. Toward Generating a New Intrusion Detection Dataset and Intrusion Traffic Characterization. ICISSP, 2018, pp. 108–116.
9. Xaitbayev A.P (2026). MOBIL QURILMALARDA PHISHING HUJUMLARINI ANIQLOVCHI DASTURIY MODUL YARATISH. Лучшие интеллектуальные исследования. 3-12 pp.
10. Chawla, N.V., Bowyer, K.W., Hall, L.O., Kegelmeyer, W.P. SMOTE: Synthetic Minority Over-sampling Technique. Journal of Artificial Intelligence Research, 2002, Vol. 16, pp. 321–357.
11. Chen, T., Guestrin, C. XGBoost: A Scalable Tree Boosting System. Proceedings of 22nd ACM SIGKDD, 2016, pp. 785–794.
12. Surayyo Baxrom qizi Rajabboyeva, Xusnutdin Kamardinovich Samarov, Azizbek Pirnazarovich Xaitbayev. (2025). AI IN THE AGE OF CYBERWARFARE: GLOBAL THREATS AND NEW DEFENSE METHODS. International Conference Platform 63-74 pp.
13. Surayyo Baxrom qizi Rajabboyeva, Xusnutdin Kamardinovich Samarov, Azizbek Pirnazarovich Xaitbayev. (2025). DEVELOPMENT OF AN INTELLIGENT MONITORING SYSTEM FOR INFORMATION SECURITY BASED ON ML. International Conference Platform. Pp 102-111.
14. Goodfellow, I., Bengio, Y., Courville, A. Deep Learning. MIT Press, 2016.
