Paper submission deadline: 26 February 99

Many people today live in information-rich worlds, constantly facing the question: what should I do next? Which papers should I read to learn about a new area I’m interested in? Which movie should I go to? Which restaurant would I like? The experience of friends and colleagues is a valuable resource for making such decisions, especially friends who are familiar with the subject area and have similar tastes.

The field of recommender systems (or collaborative filtering) attempts to automate this process, e.g., by supporting people in making recommendations, finding a set of people who are likely to provide good recommendations for a given person, or deriving recommendations from implicit behavior such as browsing activity, buying patterns, and time on task.

Within the workshop the following problems will be discussed:

• When are recommender systems appropriate? For which domains, which types of decisions, and which communities of users?

• How well do recommender systems work? Which aspects work and which are problematic?
• What is the role of persons who are interacting with a recommendation system? Do they have to provide explicit information (such as rating items) or can interest profiles be constructed automatically (e.g., by making inferences based on documents they have read, saved, deleted, etc.)? What are the pros and cons of different user and system roles?
• What types of incentives must be provided to motivate users to participate? A classic problem for rating-based systems is that early raters of any item receive no benefit, yet someone must be the first to rate in order to get the ball rolling. Various economic schemes have been proposed to address this problem—how well do they work?
• What is the role of a recommender system? Should it simply provide recommendations upon request, or can it play a more active role? How do recommender systems fit into the larger context of a person’s interaction with an application? How tightly must they be integrated with an application?

• How should recommender systems communicate their results to users? How can a user understand why a particular recommendation was offered? Should multiple recommendations be offered? Visualization techniques are of interest here.
• Who owns an individual’s recommendation data? As a user, how do I find out what use will be made of my ratings and recommendations?

The purpose of the workshop is to bring together a diverse group of researchers and practitioners from diverse fields of HCI and related disciplines, such as artificial intelligence and the social sciences. We will draw an overall picture of the state of the art and identify the most important issues facing us today. Finally, we wish to explore new ideas for recommender systems and clarify the nature, scope, and limits of this approach. One goal will be to come up with a common model of understanding, i.e. a design space in which remaining white spots can be pointed out.

1. Konstan J.A., Riedl, J., Borchers, A. Herlocker, J.L. Recommender Systems: A GroupLens Perspective, AAAI Workshop Recommender Systems 98, Papers from the 1998 Workshop Technical report WS-98-08, AAAI Press, Menlo Park California, 1998
2. The March 1997 issue of Communications of the ACM, edited by Hal Varian and Paul Resnick.
3. The Collaborative Filtering Resource web page, at http://www.sims.berkeley.edu/resources/collab

Duco Das
Philips Research Laboratories Eindhoven
das@natlab.research.philips.com