Currently, one of the most extensive issues in the field of information security is the organization of user access control to information infrastructure objects. Taking into account the volume of corporate information resources, as well as the number of users requesting access, there is a need to automate the access approval process taking into account possible risks. In this case, the most optimal solution to this problem is the use of fuzzy logic. The article analyzes the process of providing access to the information infrastructure using a fuzzy classifier and develops a conceptual model of the fuzzy classifier algorithm for incoming requests for access in order to automate the process and minimize information security risks associated with possible destructive actions aimed at the confidentiality, integrity and availability of the information infrastructure.
Keywords: neural network, machine learning, information security, cybersecurity, properties and structure of a neural network, mathematical model, threats and information vulnerabilities