Introduction. In the modern world, problems related to the use and spread of malicious software, information attacks and other types of cyber threats, which have received the general name "cybercrime" are becoming more and more relevant.During its development, the information technology sector has accumulated various types of cybercrime, which causes a great damage to both companies and individuals. According to the ISTR report [1] provided by Symantec (one of the leading developers of information security software), the past 2017 was too active for the attackers and was marked by significant incidents in Europe, the United States and the Middle East. The harm that it caused significantly exceeded the figures for 2012, when the total loss inflicted by IT offenders amounted to $ 388 billion.It is clear that IT specialists were first to realize that there were some problems with the fight against cybercrime. According to the survey, most incidents in the field of information security lead to a loss of payment data (13 %), intellectual property (13 %), customer bases (12 %) and staff information (12 %) [2]. Of course, the problem of improving the methods for analyzing network security and preventing violations in order to fight cybercrime remains relevant. Thus, in today's society, cybersecurity issues have become the defining task of protecting the global information space.Analysis of recent studies and publications. Traditional approaches to detecting malware are either limited to the use of signatures -byte sequences that identify malicious software, or heuristic algorithms, but these methods are not capable of detecting new attacks in real time [3].These days, content analysis of text information is used to prevent threats, along with the analysis of the network traffic characteristics, the behavior of corporate networks and their security policy. Existing systems of text analysis and modeling include different kinds of search engines and information-analytical systems. They are capable of solving such tasks as classification of documents by its subject matter, author identification, detection of plagiarism, modeling representations of the knowledge about the subject area and the content of text, classification and filtering of documents by specified queries, and much more [4; 5; 6].
116Highlighting the previously unsolved parts of the problem. Enhancement in the effectiveness of security systems and reduced time of threats detection requires a further development in the methods of processing the data arrays of the network packets' body, content of Internet pages and information from mass media and social networks, which raises the problem of semantic and syntactic processing of text, written in natural language.The purpose of this paper. Applying a wider range of information for assessment of cyber threat's level of danger and creation of collective protection for corporate networks through introduction of threat monitoring systems and active intelligence assessment in the global information space of the Int...
