IJCAI'99 Workshop on Learning about Users, Saturday July 31, 1999


NOTE
These on-line proceedings were re-created from archived data on September 13, 2004.

Click here to go directly to papers

Organizing Committee

Åsa Rudström (chair), Swedish Institute of Computer Science (SICS), Sweden
Dr. Mathias Bauer, German Research Center for AI (DFKI), Germany
Dr. Wayne Iba, Computational Learning Laboratory, Stanford University, USA
Dr. Wolfgang Pohl, GMD FIT, HCI Department, Germany

Program Committee

Haym Hirsh, Rutgers University, USA
Henry Lieberman, MIT Media Lab, USA
Katharina Morik, University of Dortmund, Germany
Hiroshi Motoda, Osaka University, Japan
Geoff Webb, Deakin University, Australia

Preface

As computer systems become more powerful and complex, our interactions with them have become more information laden and, consequently, more burdensome. It is now generally recognized within the HCI and intelligent user interfaces communities that as systems become more complex, this need for higher-bandwidth interfaces should be addressed by learning about and adapting to the user. The pieces to this puzzle are coming together from a variety of disciplines, including machine learning, user modeling, intelligent tutoring, information retrieval, and data mining. Furthermore, related work is discussed in the field of autonomous agents. This workshop aims at bringing together researchers from these different communities.

The goal of the workshop is to make a first step towards a framework within which research on systems that adapt to their users can be proposed, identified, conducted and evaluated. In the call for papers, we identified the following technical issues to be discussed in submitted papers and during the workshop itself:

Problem Domain:
What is the task?
What is unique about the task and why is it important?
What will a solution in this domain tell us about general solutions?
Approach or Method
How was data (from which to learn) collected?
What learning algorithm was used?
Was learning on-line or off-line?
How was the learned model utilized?
Evaluation
How do you measure success on the overall task?
How do you measure improvement for a given user?
What were the causes for success or failure?

We accepted 12 papers that are presented in these proceedings. On-line proceedings and other information about the workshop is available from the workshop web site at http://www.sics.se/humle/ijcai99-ws/.

Table of contents