Mathematical Methods in Cyber Security: Cluster Analysis and its Application in Information and Cybernetic Security

Шевченко, Світлана Миколаївна and Жданова, Юлія Дмитрівна and Спасітєлєва, Світлана Олексіївна and Мазур, Наталія Петрівна and Складанний, Павло Миколайович and Негоденко, Віталій Петрович (2024) Mathematical Methods in Cyber Security: Cluster Analysis and its Application in Information and Cybernetic Security Електронне фахове наукове видання «Кібербезпека: освіта, наука, техніка», 23 (3). pp. 258-273. ISSN 2663-4023

Text Shevchenko_S_Zhdanovа_Y_Spasiteleva_S_Mazur_N._Skladannyi_P_Nehodenko, V_cest_23_2024.pdf Download (1MB)

Abstract

The huge number of information threats and their complexity prompts research and modeling of new methodologies and information protection systems. The development and improvement of information and cyber security systems includes the creation and processing of mathematical models using information technologies. This article is a follow-up study on the application of mathematical methods and technologies in cyber security, namely: methods of cluster analysis. The modern development of computer technology and the growth of their power have contributed to the wide implementation of Data Mining algorithms for processing large volumes of information in various fields of society and science, in particular in the field of cyber security. Cluster analysis allows the set to be divided into subsets, so that the elements of each subset are similar to each other, and the elements of different subsets are the most different. This provides an opportunity to eliminate the shortcomings of the qualitative approach in assessing information risks. The paper reviews scientific sources regarding the applied aspect of the application of clustering methods in security systems, because timely forecasting of possible incidents allows you to manage information risks and make effective decisions to ensure confidentiality, availability and integrity of information. The stages of the clustering procedure are characterized, the issues of choosing the distance measure and the similarity measure for the objects under study are highlighted. The comparative characteristics of the most popular methods of cluster analysis are presented: the “nearest neighbor” algorithm, “k-means”, “fuzzy c-means”, “cosine similarity”, their advantages and disadvantages are defined. This study can be useful and used in the educational process of students of the specialty 125 “Cyber security and information protection”.

Item Type:	Article
Uncontrolled Keywords:	mathematical methods; cluster analysis; informational security; cyber security; “nearest neighbor” algorithm, “k-means” algorithm, “fuzzy c-means” algorithm, “cosine similarity” algorithm
Subjects:	Статті у періодичних виданнях > Фахові (входять до переліку фахових, затверджений МОН)
Divisions:	Факультет інформаційних технологій та математики > Кафедра інформаційної та кібернетичної безпеки ім. професора Володимира Бурячка
Depositing User:	Павло Миколайович Складанний
Date Deposited:	29 Mar 2024 12:25
Last Modified:	29 Mar 2024 12:25
URI:	https://elibrary.kubg.edu.ua/id/eprint/48471

Actions (login required)

View Item