Approach to identifying significant features of IoT device network activity | Vestnik Tomskogo gosudarstvennogo universiteta. Upravlenie, vychislitelnaja tehnika i informatika – Tomsk State University Journal of Control and Computer Science. 2025. № 73. DOI: 10.17223/19988605/73/12

Approach to identifying significant features of IoT device network activity

The study investigates the features of network activity in Internet of Things (IoT) devices and proposes a method for reducing the dimensionality of the feature space to enhance data analysis efficiency. The proposed approach eliminates multicollinearity, nonlinear dependencies, and feature redundancy while preserving their semantic interpretability. It is based on the combined use of statistical characteristics such as mutual information, correlation, stability criteria, and significance measures for feature filtering. Applying this approach significantly reduced the feature space and improved its properties: numerical stability of the data, generalization ability of models, clustering quality. Contribution of the authors: the authors contributed equally to this article. The authors declare no conflicts of interests.

Download file

Counter downloads: 2

Keywords

Internet of Things, threat datasets, statistical analysis, feature stability, feature weights in principal component analysis

Authors

Name	Organization	E-mail
Isaeva Olga S.	Institute of Computational Modelling SB RAS	isaeva@icm.krasn.ru
Isaev Sergey V.	Institute of Computational Modelling SB RAS	si@icm.krasn.ru
Kulyasov Nikita V.	Institute of Computational Modelling SB RAS	razor@icm.krasn.ru

Всего: 3

References

Курбатов В.И., Папа О.М. Интернет вещей: основные концепции и тренды // Гуманитарные, социально-экономические и общественные науки. 2023. № 1. C. 48-54.

Минаев В.А., Швырев Б.А., Ромашкин Т.Р. Безопасность интернета вещей: основные решения // Информация и безопасность. 2023. Т. 26 (2). С. 163-168.

Chatterjee A., Ahmed B.S. IoT anomaly detection methods and applications: A survey // Internet of Things. 2022. V. 19. Art. 100568. doi: 10.1016/j.iot.2022.100568.

Ring M., Wunderlich S., Scheming D., Landes D., Hotho A. A survey of network-based intrusion detection data sets // Computers & Security. 2019. V. 86. P. 147-167. doi: 10.1016/j.cose.2019.06.005.

Xinpeng С. CICIDS2017 and UNBSW-NB15 // IEEE Dataport. 2023. doi: 10.21227/ykpn-jx78. URL: https://ieee-dataport.org/documents/cicids2017-and-unbsw-nb15.

Liu L., Wang P., Lin J., Liu L.Intrusion detection of imbalanced network traffic based on machine learning and deep learning // IEEE Access. 2021. V. 9. P. 7550-7563. doi: 10.1109/ACCESS.2020.3048198.

Moustafa N. A new distributed architecture for evaluating AI-based security systems at the edge: Network TON_IoT datasets // Sustainable Cities and Society. 2021. V. 72. Art. 102994.

Горюнов М.Н., Мацкевич А.Г., Рыболовлев Д.А. Синтез модели машинного обучения для обнаружения компьютерных атак на основе набора данных CICIDS2017 // Труды Института системного программирования РАН. 2020. Т. 32 (5). С. 81-94. doi: 10.15514/ISPRAS-2020-32(5)-6.

Исаева О.С., Кулясов Н.В., Исаев С.В. Инфраструктура сбора данных и имитации угроз безопасности сети интернета вещей // Сибирский аэрокосмический журнал. 2025. Т. 26, № 1. С. 8-20. doi: 10.31772/2712-8970-2025-26-1-8-20.

Altarabichi M.G., Nowaczyk S., Pashami S., Mashhadi P.S. Surrogate-assisted genetic algorithm for wrapper feature selection // IEEE Congress on Evolutionary Computation. 2021. P. 776-785. doi: 10.1109/CEC45853.2021.9504718.

Liu S., Motani M. Improving mutual information based feature selection by boosting unique relevance // Journal of Artificial Intelligence Research. 2025. V. 82. P. 1267-1292. doi: 10.1613/jair.1.17219.

Исаева О.С. Построение цифрового профиля устройств Интернета вещей // Информационные и математические технологии в науке и управлении. 2023. № 2 (30). С. 36-44. doi: 10.25729/ESI.2023.30.2.004.

Shafi M., Lashkari A., Roudsari A. NTLFlowLyzer: Towards generating an intrusion detection dataset and intruders behavior profiling through network and transport layers traffic analysis and pattern extraction // Computers & Security. 2025. V. 148. Art. 104160. doi: 10.1016/j.cose.2024.104160.

Исаева О.С. Построение онтологии для систематизации характеристик сети Интернета вещей // Онтология проектирования. 2024. Т. 14, № 2 (52). С. 243-255. doi: 10.18287/2223- 9537-2024-14-2-243-255.

Priyatno A.M., Widiyaningtyas T. A systematic literature review: recursive feature elimination algorithms // JITK. 2024. V. 9 (2). P. 196-207. doi: 10.33480/jitk.v9i2.5015.

Цурко В.В., Михальский А.И. Оценка статистической связи случайных величин через взаимную информацию // Автоматика и телемеханика. 2022. Вып. 5. C. 76-86. doi: 10.31857/S0005231022050063.

Koltchinskii V. Rademacher complexities and bounding the excess risk // Journal of Machine Learning Research. 2010. V. 11. P. 2457-2485.