Assessment of Current Threats to Information Security Using Transformer Technology

Abstract

Purpose: development of an automated system for assessing current threats to the security of software of industrial automation systems based on the technology of Transformers. Methods: comparison of the set of identified software vulnerabilities, corresponding tactics (techniques) and relevant threats to information security by assessing the semantic proximity metrics of their text descriptions using Text Mining technology based on transformers models. Practical relevance: an automated system for assessing current software security threats has been developed, which makes it possible to compare and rank information and cyber security threats for identified vulnerabilities from the FSTEC of Russia Information Security Threats Databank, to automate the selection of techniques and tactics for constructing threat scenarios. The results of the comparative analysis show that the use of this system makes it possible to simplify the procedure for selecting potential threats and comparing vulnerabilities to them, in addition, a possible set of tactics and techniques is automatically generated, which makes it possible to reduce the time spent on building scenarios for the implementation of threats.