Scrutable Personalisation

Associate Professor Judy Kay
School of Information Technologies
The University of Sydney

Tuesday 21st January at 11am

Abstract

There is a growing interest in building personalised systems. These adapt their interaction on the basis of a user model. This represents the system's beliefs about the user. This seminar explains the motivation for scrutability in such personalised systems. It also presents a brief overview of the accretion representation. This was motivated by the goal of scrutability of the user model. It supports user modelling where users can see what the system believes about them. It further enables users to examine the processes that contributed to the system's beliefs in the user model.

The seminar then describes a selection of current projects which build on this basic work. First, I will describe and give a brief demonstration of an innovative visualisation interfaces which enable the user to gain an overview of their user model. I will briefly describe how this is being enhanced with ontological inferences. Next, I will give an overview and demonstration of a system for providing scrutably personalised hypertext. Then, there will be a very brief overview of a project to support reuse of learning document (learning objects in the current vogue), a medical teaching systems that encourages reflection based on the system's model of this user and others and an even briefer overview of some of the other activities related to supporting scrutable personalisation.

Short resume

 

Judy Kay is an Associate Professor at the School of Information Technologies (formerly, Basser Department of Computer Science) at the University of Sydney. She is a principal in the Smart Internet Technology Research Group which conducts fundamental as well as applied research in user-adapted systems, based on customisation and delivery of multi-media objects. Projects focus on scrutable user modelling modelling and the systems aspects of managing and delivering user model objects as well the multimedia objects users want. The areas in which she leads research include intelligent teaching systems and customising and filtering of information. The group has significant deployed research, including a user-based CPU scheduling system, the FairShare Scheduler. Recent major grants to the group include the Smart Internet Technology CRC and Science Lectureships - Building the Internet Workforce.

She has over 100 publications in the areas of personalisation and teaching and learning. These appear in top conferences and journals in this area: the User Modeling conferences; the User Modeling and User-Adapted Interaction journal; and Communications of the ACM. This work has led to the invited keynote addresses an major conferences: UM'94 User Modeling Conference; IJCAI'95 International Joint Conference on Artificial Intelligence; ICCE'97, International Conference on Computers in Education; and ITS'2000, Intelligent Tutoring Systems.

Back to HAIL Home Page