AI-DRIVEN THREAT DETECTION AND PREVENTION IN AI
DOI:
https://doi.org/10.65164/51wr8a40Ключевые слова:
artificial intelligence, machine learning, threat detection, cybersecurity, intrusion detection system, deep learning, anomaly detection, adversarial attacks, zero-day exploits, SIEM.Аннотация
The rapid proliferation of sophisticated cyber threats has rendered traditional signature-based security measures increasingly inadequate. This thesis explores the application of artificial intelligence (AI) and machine learning (ML) techniques for automated threat detection and prevention in modern cybersecurity environments. The study examines supervised, unsupervised, and deep learning methodologies, evaluating their effectiveness across various attack vectors including intrusion detection, malware classification, phishing identification, and zero-day exploit mitigation. Experimental findings demonstrate that ensemble deep learning architectures achieve detection accuracy exceeding 96%, significantly outperforming conventional methods. The paper further discusses challenges of adversarial attacks against AI systems and proposes a hybrid adaptive framework combining real-time behavioral analytics with threat intelligence feeds. Results indicate that AI-powered cybersecurity systems can substantially reduce mean time to detect (MTTD) and mean time to respond (MTTR), offering a viable pathway toward autonomous, self-healing cyber defense infrastructures.
Библиографические ссылки
[1] IBM Security. Cost of a Data Breach Report 2023. – IBM Corporation, Armonk, NY, 2023. – 74 p.
[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. – P. 1153–1176.
[3] Sarker I.H., Kayes A.S.M., Badsha S. et al. Cybersecurity Data Science: An Overview from Machine Learning Perspective // Journal of Big Data. – 2021. – Vol. 8. – Article 37. DOI: 10.1186/s40537-020-00318-5.
[4] Rakhimov M.A., Yusupov R.I. Applying Machine Learning for Cybersecurity Threat Detection in Central Asian Network Infrastructures // Vestnik TUIT (Bulletin of TUIT). – 2022. – No. 3(59). – P. 44–53.
[5] Kotenko I.V., Chechulin A.A. Mnogoagentnyye sistemy dlya intellektualnoy zashchity informatsionnykh setey [Multi-agent Systems for Intelligent Protection of Information Networks] // Trudy SPIIRAN. – 2020. – Vol. 19, No. 6. – P. 1267–1298.
[6] Chen Z., Liu J., Shen Y. et al. BERT-Based Log Anomaly Detection for Cybersecurity Threat Identification // Computers and Security. – 2023. – Vol. 128. – Article 103146. DOI: 10.1016/j.cose.2023.103146.
[7] Li Y., Ma R., Jiao R. A Hybrid Malicious Code Detection Method Based on Deep Learning // International Journal of Security and Its Applications. – 2022. – Vol. 9, No. 5. – P. 205–216.
[8] Apruzzese G., Colajanni M., Ferretti L. et al. Addressing Adversarial Attacks Against Security Systems Based on Machine Learning // IEEE Access. – 2022. – Vol. 10. – P. 15184–15199.
[9] Khraisat A., Gondal I., Vamplew P., Kamruzzaman J. Survey of Intrusion Detection Systems: Techniques, Datasets and Challenges // Cybersecurity. – 2023. – Vol. 6. – Article 20. DOI: 10.1186/s42400-023-00132-x.
[10] Qodirov B.S., Toshmatov S.A. Sun'iy intellekt asosida kiberxavfsizlik tizimlarini loyihalash [Designing Cybersecurity Systems Based on Artificial Intelligence] // O'zbekiston Respublikasi Fanlar Akademiyasining Ma'ruzalari. – 2023. – No. 2. – P. 78–85.
[11] Apruzzese G., Laskov P., Schneider J. SoK: The Impact of Unlabelled Data in Cyberthreat Detection // Proceedings of the IEEE Symposium on Security and Privacy (SP). – 2023. – P. 1218–1235.
[12] Mirzoev T.B. Kibertahdidlarni aniqlash va bartaraf etishda mashinaviy o'qitish usullari [Machine Learning Methods in Detection and Mitigation of Cyber Threats] // Axborot Texnologiyalari va Telekommunikatsiyalar. – 2022. – No. 4(16). – P. 12–20.
