N.V. Gololobov, E.Yu. Pavlenko, D.S. Lavrova

The investigated scientific and technical problem is related to the growing number of cyberattacks using malicious software (malware). Such attacks represent a complex security threat due to the wide range of possible destructive impacts affecting target computer systems. Malware is characterized by stealthy action through masking of malicious payloads or mimicry as legitimate software, which complicates their detection and neutralization.

The research results consist of obtaining knowledge about the peculiarities of behavioral characteristics of malware and their influence on telemetry data, as well as expanding the methodological basis for security threat analysis through the use of indirect behavioral indicators. The validation results demonstrate the feasibility of using an architecture-invariant software solution that operates on reliable data from hardware components. Experimental studies confirmed the viability of the method: the best detection accuracy was 100%, classification accuracy was 90%. Due to the use of machine learning technology, the method can be retrained on new malware samples to maintain relevance.

Gololobov N.V., Pavlenko E.Yu., Lavrova D.S. A method for intelligent classification of malicious software based on telemetry of computer hardware components. Radiotekhnika. 2026. V. 90. № 2. P. 57−65. DOI: https://doi.org/10.18127/j00338486-202602-08 (In Russian)

1. Kim S.K. Advanced machine learning based malware detection systems. IEEE Access. 2024. V. 12. P. 45721–45735. DOI: 10.1109/ACCESS.2024.10613413.
2. Chao J., Xie T. Deep learning-based network security threat detection and defense. International Journal of Advanced Computer Science and Applications. 2024. V. 15. № 11. DOI: 10.14569/IJACSA.2024.0151164. 
3. Saravanan V., Khushboo Tripathi, Kurivella N.S.K Santhosh, Naveenkumar Palanivel, Vidyasri P. AI-Driven cybersecurity: enhancing threat detection and mitigation with deep learning. International Journal of Computational and Experimental Science and Engineering. 2025. V. 11. № 2. DOI: 10.22399/ijcesen.1358. 
4. University of New Brunswick. CIC IDS 2017 Dataset. Rezhim dostupa: https://www.unb.ca/cic/datasets/ids-2017.html. 
5. Kaggle. CERT Insider Threat Dataset. Rezhim dostupa: https://www.kaggle.com/datasets/nitishabharathi/cert-insider-threat. 
6. PhishTank. Developer Info. Rezhim dostupa: https://www.phishtank.com/developer_info.php. 
7. Wang K. Leveraging AI for automated malware classification and detection. Journal of Information Systems Engineering and Management. 2025. V. 10. № 1. 
8. Higuera J.B., Lima V.P., Gómez R. Semantic malware classification using artificial intelligence. Computer Modeling in Engineering & Sciences. 2025. V. 142. № 3. P. 3031–3067. DOI: 10.32604/cmes.2025.061080. 
9. Saini V., Gupta R., Soni N. OpCode-based malware classification using machine learning and deep learning techniques. arXiv preprint arXiv:2504.13408. 2025. DOI: 10.48550/arXiv.2504.13408. 
10. McLaughlin N., Martinez del Rincon J., Kang B., Yerima S. Deep Android Malware Detection. Proceedings of the 32nd Annual ACM Symposium on Applied Computing. 2017. DOI: 10.1145/3029806.3029823. 
11. Positive Technologies. Malware behavior and distribution channels report. [Jelektronnyj resurs]. URL: https://ptsecurity.com/ru-ru/research/analytics/malware-behavior-and-distribution-channels/#id2.
12. Gololobov N.V. Sistematizacija vredonosnogo programmnogo obespechenija dlja opredelenija tipov projavljaemoj imi aktivnosti. Problemy informacionnoj bezopasnosti. Komp'juternye sistemy. 2024. № 3(61). S. 142–154. DOI: 10.48612/jisp/npp7-45tr-em48 (in Russian).
13. Yunjia W. et al. Functional dependency discovery with the FDX algorithm. Proceedings of the ACM SIGMOD Conference. 2020.
14. Karpuk V.A. Vydelenie funkcional'nyh zavisimostej v reljacionnoj baze dannyh. Minsk: BGUIR. 2020 (in Russian).
15. Papenbrock T., Naumann F.  Hybrid approach to discover functional dependencies. Proceedings of the VLDB Endowment. 2016. V. 8. № 10. P. 1082–1093.
16. Simonenko E.Ju. Problema vyjavlenija funkcional'nyh zavisimostej v reljacionnoj BD. Nauchno-tehnicheskij vestnik informacionnyh tehnologij, mehaniki i optiki. 2007. S. 109-113 (in Russian).