MODERN APPROACHES TO THE ANALYSIS OF TEXTUAL INFORMATION ABOUT SECURITY EVENTS BASED ON ARTIFICIAL INTELLIGENCE METHODS

Rusakov Alexey Mikhailovich, Komarov Kirill Yurievich, Koryagin Sergey Viktorovich, Koryagina Veronika Mikhailovna

In the context of the rapid growth of text information volumes, intelligent data analysis is becoming a key tool for processing and structuring information about vulnerabilities of information systems. This paper discusses methods for automating the analysis of text descriptions of vulnerabilities using artificial intelligence and machine learning technologies. The relevance of the study is due to the growth of cyber threats, especially after the COVID-19 pandemic, when the mass transition to remote work led to an increase in the number of attacks on government and commercial structures. In Russia, a number of regulations have been adopted to counter these threats, including Federal Law No. 187-FZ and FSTEC methods, but vulnerability analysis still requires significant time due to manual data processing. The article provides an overview of modern vulnerability description systems (CVE, CWE, NVD), methods for their classification, as well as approaches to automated processing of text data using software libraries and machine learning algorithms, and attention is also paid to reducing the size of the data and visualizing the results. The results of the study show that the use of text mining allows for faster and more accurate processing of vulnerability descriptions, which contributes to more effective cyber risk management.

Keywords:information security events, text mining, artificial intelligence methods