This tutorial deals with the broad class of systems that adapt to properties and behaviors of individual users. These systems perform many different functions, including information retrieval and presentation, product recommendation, instruction, intelligent help, interface adaptation, automated execution of the user's routine tasks, and support for collaboration. The AI techniques used include machine learning methods, probabilistic and decision-theoretic approaches, logic-based techniques, and various more or less application-specific methods.
Research on user-adaptive systems has been unnecessarily fragmented, with specialized communities focusing on particular types of application or particular implementation techniques. This tutorial aims to further cross-fertilization among these communities by providing a unifying conceptual framework as well as pointers to major developments throughout the area.
The tutorial will consider in turn six questions that can be asked about any user-adaptive system:
For each question, we will compare and evaluate the contributions that have been made so far. The entire presentation will be illustrated with concrete examples of existing systems.
This tutorial will be accessible to persons with no previous knowledge of user-adaptive systems. It is also aimed at experienced researchers in this area who want to make better use of relevant research that has been conducted outside of their own research communities.
is a senior project director and lecturer at the
University of Saarbruecken who also participates in industrial
projects at the German Research Center for AI (DFKI). He has published
on various aspects of user-adaptive systems since the early 1980's and
was program co-chair of UM97, the Sixth International Conference on