Keywords
Abstract
Ushbu maqolada korporativ tarmoqlarda uchraydigan kiberhujumlarni erta aniqlash jarayonida chuqur o‘rganish (Deep Learning) algoritmlaridan samarali foydalanish masalasi ilmiy asosda tahlil qilinadi. So‘nggi yillarda korporativ infratuzilmalarda tarmoq trafikining hajmi ortishi, kiber tahdidlarning murakkablashuvi va Zero-Day turdagi hujumlarning ko‘payishi xavfsizlik tizimlaridan yuqori darajadagi intellektual yondashuvlarni talab qilmoqda. Shu nuqtai nazardan maqolada konvolyutsion neyron tarmoqlar (CNN), rekurrent neyron tarmoqlar (RNN, LSTM), avtomatik kodlovchilar (Autoencoders) kabi chuqur o‘rganish modellarining kiberhujumlarni aniqlashdagi afzalliklari, real vaqt rejimidagi ishlash ko‘rsatkichlari va anomaliya tahlilidagi aniqligi o‘rganiladi. Tadqiqotda korporativ tarmoq trafikidan olingan ma’lumotlar asosida modellarni o‘qitish, ularning klassifikatsiya aniqligi, noto‘g‘ri pozitiv va noto‘g‘ri negativ ko‘rsatkichlari ilmiy tahlil qilinadi.
References
1. Ahmadov, R. (2022). Korporativ tarmoqlarda axborot xavfsizligini ta’minlashning zamonaviy yondashuvlari. Toshkent: Innovatsion texnologiyalar nashriyoti.
2. CICIDS Dataset. (2017). Canadian Institute for Cybersecurity. https://www.unb.ca/cic/datasets/
3. Gao, J., Wang, Y., & Chen, F. (2021). Deep learning-based network traffic classification and anomaly detection. IEEE Access, 9, 92333–92345.
4. Kim, J., Lee, H., & Kim, J. (2020). CNN–LSTM hybrid deep learning model for intrusion detection in software-defined networks. Security and Communication Networks, 2020, 1–12.
5. Mahmood, Z., & Afzal, M. (2021). Zero-Day attack detection using unsupervised deep learning models. Journal of Network and Computer Applications, 178, 102981.
6. NSL-KDD Dataset. (2015). University of New Brunswick. https://www.unb.ca/cic/datasets/nsl.html
7. Sharif, A., Raza, M., & Shah, M. (2020). Anomaly-based intrusion detection using autoencoders in large-scale enterprise networks. Computers & Security, 98, 101993.
8. Yusupov, A. (2023). Kiberhujumlarga qarshi tizimlarda mashinaviy o‘rganish modellarining qo‘llanilishi. Toshkent: O‘zbekiston Milliy Universiteti nashriyoti.
9. Zhang, Y., Qin, Z., & Lin, X. (2019). Deep learning-based cybersecurity techniques for network intrusion detection: A comprehensive review. IEEE Access, 7, 172231–172250.