DOI
10.34229/KCA2522-9664.25.6.3
UDC 004.05
O.O. Letychevskyi
V.M. Glushkov Institute of Cybernetics, National Academy of Sciences of Ukraine,
Kyiv, Ukraine, 
oleksandr.letychevskyi@litsoft.com.ua
NEURO-SYMBOLIC TWINS OF SYSTEMS AND THEIR USE
IN SOLVING CYBERSECURITY PROBLEMS
Abstract. The paper considers the use of software system twins, which are presented as a combination of an algebraic model and a neural network. The use of this technology in intrusion detection systems and at the stage of assessing the reliability of cyber defense of software systems is considered. The considered approach can significantly increase the accuracy of detecting cyber attacks in real-time, counteracting adversarial attacks, and avoiding false-positive detections, which is a problem in detecting attacks with unknown semantics. The use of a neuro-symbolic twin at the stage of preparedness system for operation, namely the procedure for identifying system vulnerabilities, is considered. The architecture of an intrusion detection system based on a neuro-symbolic twin, which can monitor the incoming communication protocol and restore the software environment, is presented. Examples of using the technology in a blockchain environment and in hardware protection are given.
Keywords: digital twin, cybersecurity, algebraic modeling, deep learning neural network, adversarial attacks, vulnerability detection, intrusion detection system.
full text
REFERENCES
- 1. Pinto A., Herrera L.-C., Donoso Y., Gutierrez J.A. Survey on intrusion detection systems based on machine learning techniques for the protection of critical infrastructure. Sensors. 2023. Vol. 23, Iss. 5. Article number 2415. https://doi.org/10.3390/s23052415.
- 2. Hitzler P., Eberhart A., Ebrahimi M., Sarker M.K., Zhou L. Neuro-symbolic approaches in artificial intelligence. National Science Review. 2022. Vol. 9, Iss. 6. Article number nwac035. https://doi.org/10.1093/nsr/nwac035.
- 3. Ansys Twin Builder. Create and deploy digital twin models. URL: https://www.ansys.com/en-gb/products/digital-twin/ansys-twin-builder.
- 4. Learn about twin models and how to define them in Azure Digital Twins. URL: https://learn.microsoft.com/en-us/azure/digital-twins/concepts-models.
- 5. Digital Twin Definition Language. URL: https://github.com/Azure/Azure/opendigitaltwins-dtdl/blob/master/DTDL/v3/DTDL.v3.md.
- 6. Letychevskyi O. Congnitive networks, their properties and applications in attack detection and prevention systems. Cybernetics and Systems Analysis. 2023. Vol. 59, N 5. P. 859–866. https://doi.org/10.1007/s10559-023-00621-y.
- 7. Letychevskyi O., Peschanenko V. Applying algebraic virtual machine to cybersecurity tasks. Proc. 2022 IEEE 9th International Conference on Sciences of Electronics, Technologies of Information and Telecommunications (SETIT) (28–30 May 2022, Hammamet, Tunisia). Hammamet, 2022. P. 161–169. https://doi.org/10.1109/SETIT54465.2022.9875895.
- 8. Goseva-Popstojanova K., Tyo J. Identification of security related bug reports via text mining using supervised and unsupervised classification. Proc. 2018 IEEE 18th International Conference on Software Quality, Reliability and Security (QRS) (16–20 July 2018, Lisbon, Portugal). Lisbon, 2018. P. 344–355. https://doi.org/10.1109/QRS.2018.00047.
- 9. Zhou Y., Liu S., Siow J., Du X., Liu Y. Devign: Effective vulnerability identification by learning comprehensive program semantics via graph neural networks. arXiv:1909. 03496v1 [cs.SE] 8 Sep 2019. https://doi.org/10.48550/arXiv.1909.03496.
- 10. Collyer J., Watson T., Phillips I. Know your neighborhood: General and zero-shot capable binary function search powered by call graphlets. arXiv:2406.02606v2 [cs.CR] 11 Nov 2024. https://doi.org/10.48550/arXiv.2406.02606.
- 11. Grieves M. Origins of the Digital Twin concept. 2016. https://doi.org/10.13140/RG.2.2.26367.61609.
- 12. Letychevskyi O.O., Peschanenko V.S., Hryniuk Y.V., Radchenko V.Yu., Yakovlev V.M. An overview of the modern methods of security and protection of software systems. Cybernetics and Systems Analysis. 2019. Vol. 55, N 5. P. 840–850. https://doi.org/10.1007/s10559-019-00194-9.
