Оригинал (Original) | |||||
---|---|---|---|---|---|
Автори: Христов, А. В. Заглавие: ИЗПОЛЗВАНЕ НА ЛОГИСТИЧНА РЕГРЕСИЯ ЗА ОТКРИВАНЕ НА КИБЕРАТАКИ В ИНТЕРНЕТ НА НЕЩАТА Ключови думи: Cyberattacks, Detection, Internet of Things, Logistic regres Абстракт: Целта на настоящата статия е като се анализират последните тенденции в откриването на мрежови атаки, насочени към Интернет на нещата, да се предложи комбиниран детектор, базиран на логистична регресия, използван за идентифициране на компрометирани устройства в Интернет на нещата - IoT. Библиография Издание
| Autors: Hristov, A. V. Title: USING LOGISTICS REGRESSION FOR DETECTING CYBERATTACKS IN INTERNET OF THINGS Keywords: Cyberattacks, Detection, Internet of Things, Logistic regres Abstract: The aim of present paper is to analyze recent trends in the detection of network attacks targeting Internet of Things devices as well as to propose a combined detector based on logistic regression and Emotional Model module used to identify compromised Internet of Things devices. The combined detector is evaluated and two numerical experiments are conducted. The achieved accuracy is approx. 0.94. The proposed detector is more accurate than other solutions proposed by other authors. For example, the random forest of decision trees algorithm (Random Forest-RF) and Dense Neural Network, achieve accuracy approx. 0.87. References Issue
|
Цитирания (Citation/s):
1. Stefanova-Stoyanova, V., I. Stankov, B. Danov, Exploring the Synergy between Zero-knowledge Proof and Smart Questioning, Proceedings of the 11th International Scientific Conference COMPUTER SCIENCE, IEEE Conference, Rec # 59259, 18 – 20 September 2023, Sozopol, Bulgaria, DOI: 10.1109/COMSCI59259.2023.10315909 - 2023 - в издания, индексирани в Scopus или Web of Science
2. 2) Andreev, D. R. Trifonov and M. Lazarova, Challenges Regarding AI Integration in V2X Communication, Proceedings of the 12th International Scientific Conference COMPUTER SCIENCE, IEEE Conference, Rec # 63166, 13 – 15 September 2024, Sozopol, Bulgaria, DOI - 2024 - в издания, индексирани в Scopus или Web of Science
Вид: статия в списание, публикация в реферирано издание