DEVELOPMENT OF A MACHINE LEARNING ALGORITHM FOR RISK ASSESSMENT IN INTERNET OF THINGS (IOT) SYSTEMS

Happy Ondobo Steve
University of Buea, Buea, Cameroon, happionodobo@gmail.com

Vitaly B. Kreindelin
MTUCI, Moscow, Russia, vitkrend@gmail.com

DOI: 10.36724/2664-066X-2026-12-4-32-38

SYNCHROINFO JOURNAL. Volume 12, Number 4 (2026). P. 32-38.

Abstract

Machine learning offers a promising approach to automating risk assessment by analyzing patterns in network traffic. This work builds upon existing anomaly and intrusion detection methods while addressing the lack of IoT-specific risk assessment frameworks that incorporate both the likelihood of occurrence and the impact of security incidents. The rapid expansion of the Internet of Things (IoT) has created significant challenges for ensuring the security and resilience of interconnected devices. Traditional security approaches are insufficient to protect against dynamic, evolving cyber threats. This paper proposes an IoT risk assessment framework based on machine learning, integrating both supervised and unsupervised techniques to identify, classify, and mitigate threats. The framework evaluates risks based on likelihood and impact, providing a scalable and adaptive solution suitable for modern IoT networks.

Keywords:  IoT security, risk assessment, machine learning, anomaly detection, cybersecurity

References

[1] OWASP Foundation. IoT Top Ten Security Threats. 2018.

[2] R. Sommer, V. Paxson, “Outside the Closed World: On Using Machine Learning for Network Intrusion Detection,” In 2010 IEEE Symposium on Security and Privacy. 2010.

[3] D. Smith, et al., “Cybersecurity Challenges in IoT: A Risk-B ased Approach,” Journal of Cybersecurity. 2023.

[4] A. Dow, “Machine Learning for IoT Security: Techniques and Applications,” IEEE Transactions on IoT. 2022.

[5] Y. Meidan, M. Bohadana, Y. Mathov, et al., “N-BaIoT – Network-Based Detection of IoT Botnet Attacks Using Deep Autoencoders,” IEEE Pervasive Computing. 2018. https://ieeexplore.ieee.org/document/8324144.

[6] S. Sicari, A. Rizzardi, L. A. Grieco, A. Coen-Porisini, “A Multidimensional Robust and Adaptive Framework for IoT Risk Assessment,” Computer Communications. 2020. https://www.sciencedirect.com/science/article/abs/pii/S0140366419318341

[7] National Institute of Standards and Technology (NIST). Considerations for Managing Internet of Things (IoT) Cybersecurity and Privacy Risks (NISTIR 8228). 2019. https://nvlpubs.nist.gov/nistpubs/ir/2019/NIST.IR.8228.pdf

[8] H. H. Pajouh, R. Javidan, R. Khayami, et al., “A Hybrid Deep Learning Model for Intelligent Threat Detection in IoT Systems,” IEEE Access. 2022. https://ieeexplore.ieee.org/document/9743117

[9] N. T. Nguyen, D. T. Hoang, D. N. Nguyen, et al., Federated Learning for IoT Security: A Case Study on Botnet Attack Detection,” IEEE Transactions on Industrial Informatics. 2023. https://ieeexplore.ieee.org/abstract/document/10020347

[10] National Institute of Standards and Technology (NIST). Cybersecurity Framework (Version 1.1). 2018.

[11] A. Alsaeedi, M. M. Rizvi, K. A. Al-Shqeer, “IoT Network Intrusion Dataset,” MDPI Data, 2019, no. 4(4), 141. https://www.mdpi.com/2306-5729/4/4/141

[12] Palo Alto Networks (Unit 42), The Role of AI and ML in IoT Security. 2023. https://unit42.paloaltonetworks.com/ai-machine-learning-iot-security/

13. V. B. Kreindelin, N. A. Legkov, “Protection of Authentication Data for Websites and Web Applications,” Telecommunications and Information Technologies. 2022. Vol. 9, No. 1, pp. 6-10.

[14] V. B. Kreindelin, A. D. Avidzba, “Encryption wi-fi protected access,” Technologies of the information society: XI International Industry Scientific and Technical Conference: collection of proceedings, Moscow, March 15-16, 2017. Moscow: Media Publisher, 2017. P. 294.

[15] V. B. Kreindelin, G. A. Vakhromeev, “Most Effective Machine Learning Algorithms for Risk-Based Authentication Systems,” Telecommunications and Information Technologies. 2024. Vol. 11, No. 1, pp. 87-92.