KIBERXAVFSIZLIKDA TAHDIDLARNI ERTA ANIQLASH USULLARINING EVOLUTSIYASI: IMZO (SIGNATURE-BASED), ANOMALIYA VA GIBRID YONDASHUVLARNING QIYOSIY TAHLILI

Samarov X. Q. & Odilov A. O.


Annotatsiya

Hozirgi shiddat bilan rivojlanib borayotgan davrda kiberxavfsizlik muhitida tahdidlar tobora murakkablashib, ko‘p bosqichli, yashirin va avtomatlashtirilgan xususiyat kasb etmoqda. Ushbu maqolada tarmoq tahdidlarini erta aniqlash sohasidagi asosiy yondashuvlar – imzo-asosli (signature-based), anomaliyaga asoslangan (anomaly-based) va gibrid (hybrid) tizimlarning evolutsion rivojlanishi hamda ularning samaradorligi qiyosiy tahlil qilinadi. Tadqiqot davomida an’anaviy imzo tizimlarining nolinchi kun (zero-day) hujumlarga qarshi ojizligi, sof mashinali o‘qitish modellarining yuqori noto‘g‘ri ogohlantirish (False Positive Rate – FPR) darajasi va gibrid yondashuvlarning ushbu cheklovlarni bartaraf etishdagi potensiali o‘rganildi. MITRE ATT&CK MITRE (Adversarial Tactics, Techniques, and Common Knowledge — bu hujumchilarning qanday hujum qilishini, qaysi bosqichlarda qanday usullardan foydalanishini tizimli ravishda tavsiflovchi bilimlar bazasidir) frameworki kontekstida o‘tkazilgan tahlillar shuni ko‘rsatadiki, deterministik filtrlash va nazoratsiz chuqur o‘qitishni birlashtirgan gibrid arxitekturalar aniqlash aniqligini (Precision ≥90%) saqlagan holda, FPR ni ≤5% ga tushirish va aniqlash vaqtini (MTTD) sezilarli darajada qisqartirish imkonini beradi. Maqola ilmiy jamoatchilikka tahdidlarga asoslangan himoya (threat-informed defense) strategiyalarini tanlashda metodologik asos bo‘lib xizmat qiladi.


Adabiyotlar

  1. IBM Security. Cost of a Data Breach Report 2024. – Armonk: IBM, 2024. – 68 p.
  2. Sommer R., Paxson V. Outside the Closed World: On Using Machine Learning for Network Intrusion Detection // IEEE Symposium on Security and Privacy. – 2010. – P. 305–316.
  3. Alazab M. et al. Machine Learning and Deep Learning Techniques for Cybersecurity: A Review // IEEE Access. – 2024. – Vol. 12. – P. 15670–15689.
  4. Apruzzese G. et al. The Role of Machine Learning in Cybersecurity: A Survey of Challenges and Opportunities // Digital Threats: Research and Practice. – 2024. – Vol. 5, №1. – P. 1–35.
  5. Chen L. et al. Hybrid Intrusion Detection System Combining Rule-based Filtering and Unsupervised Deep Learning // Journal of Information Security and Applications. – 2025. – Vol. 89. – P. 103–115.
  6. SigmaHQ. Sigma Rule Format Specification v1.0. – Open Source Project, 2025. – URL: https://github.com/SigmaHQ/sigma (murojaat sanasi: 18.01.2026).
  7. MITRE Corporation. ATT&CK Framework Version 18.0: Tactics, Techniques & Procedures. – Bedford: MITRE Engenuity, 2025. – URL: https://attack.mitre.org (murojaat sanasi: 10.01.2026).
  8. Nguyen T. et al. Adaptive Risk Scoring for Cybersecurity Alerts Using MITRE ATT&CK Context // Computers & Security. – 2025. – Vol. 151. – P. 104–120.
  9. ENISA. Artificial Intelligence in Cybersecurity: Opportunities and Challenges. – Heraklion: European Union Agency for Cybersecurity, 2024. – 74 p.
  10. NIST. AI Risk Management Framework (AI RMF 1.0). – Gaithersburg: National Institute of Standards and Technology, 2023. – 56 p.