Automatic reconstruction of the character and temperament of users based on multi-agent training of neurocognitive models of the conscious and unconscious according to data on user behavior in the internet

The purpose of this study is to develop a system for teaching simulation models of users to the properties of character and temperament based on data on their behavior in the Internet. To model user behavior, the metaphor of designing an intelligent software agent is used; the agent is controlled on the basis of a multi-agent neurocognitive architecture, which is well suited for teaching specialized patterns of behavior of specific users, information about which can be collected in the Internet. As a result of the study, the basic principles of imitation modeling of functional systems of unconscious and conscious cognitive processes were developed based on the invariant of a recursive multi-agent neurocognitive architecture. Algorithms for the simulation of functional systems of character and temperament based on the invariant of a recursive multi-agent neurocognitive architecture have been substantiated. The results obtained can be used to develop basic principles, models, methods and algorithms for learning the invariant of a multi-agent neurocognitive architecture based on data on user behavior and mentions about it in the Internet space.