With the advances in communication technology, large volumes of information can transfer across continents within a fraction of a second. Nevertheless, computer users still suffer from unpleasant situations when they interact with systems and are required to adapt to systems rather than the other way round. User modeling aims to overcome this problem by enabling computer systems to interact with users according to the users’ models, i.e., goals, knowledge, and preferences of users. Although, user modeling has shown invaluable benefits, methods of capturing user information to build precise and useful user models are still in their early states. This paper proposes a new approach for gathering user information by pooling the information from different systems. This concept, entitled Distributed User Modeling (DUM) is based on a method in which sensors built into each system contribute specific user information to the pooling. Having multiple sources of user information increases the possibility that a system can generate reliable user models. A general model of DUM is presented in this paper. The conceptual framework of the Distributed Fuzzy Object-Oriented User Modeling System (DFOOUMS) that uses DUM as its basis structure is also presented.
Piyawat, Thawatchai; Adya, Monica; and Norico, Anthony F., "The Distributed User Modeling Shell System (DUMSS): A Conceptual Framework for Eliciting User Models" (2000). AMCIS 2000 Proceedings. 330.