Machine Learning for Analyzing and Modeling the Behavior of Computer System Users

The authors present an overview of the current state and the main scientific results of the team of the Department of Intellectual Information Technologies of the Faculty of Computational Mathematics and Cybernetics of the Moscow State University in the field of research and development of machine learning methods for solving problems of analysis and modeling of the behavior of users of computer systems. Models and problems for major sources of behavioral data, including the human-machine interface, application and system logs, electronic documents, and interacting user groups and communities, as well as combinations of these sources, are considered. Emphasis is placed on solving problems related to computer and information security.