Ontology-oriented user clustering based on the history of interaction between user and knowledge management system

Knowledge management is a combination of the processes which manage creation, extraction, processing, utilization, and distribution of knowledge in a certain domain. The knowledge management system is a complex of the procedures which realize these processes. Ontologies are widely used today for the knowledge representation both in Russia and abroad. A part of any system interacting with user is a possibility to personify the information and knowledge flow between the system and the user.
In this paper the method of ontology-oriented clustering for grouping knowledge management system users based on their preferences is proposed. The knowledge-based methods of artificial intelligence, profiling and clustering are used in this paper.
Described in the paper method of ontology-oriented user clustering allows to identify groups of users and common preferences of these groups. It allows adapting information and knowledge flow between user and system based on these preferences. Input data for this method is a user requests and ontology slices which corresponded to these requests. This information is keeping in the user profile. The complexity of this method is $O(N^3)$, $N$ — count of user requests, for the large amount of user requests.
This method can be widely applied in the user-oriented knowledge-based systems. For example it can be applied in expert systems, intellectual searching systems, knowledge management systems for production networks and other. Personification of such systems is very important task in modern life, because it allows automating the processes of interaction system with user and interaction between users in these systems.