Toward CaseBased Preference Elicitation: Similarity Measures on Preference Structures
Vu Ha Peter Haddawy Decision Systems and Artificial Intelligence Lab Dept. of EE&CS University of WisconsinMilwaukee Milwaukee, WI 53201
While decision theory provides an appealing normative framework for representing rich preference structures, eliciting utility or value functions typically incurs a large cost. For many applications involving interactive sys tems this overhead precludes the use of for mal decisiontheoretic models of preference. Instead of performing elicitation in a vacuum, it would be useful if we could augment di rectly elicited preferences with some appro priate default information. In this paper we propose a casebased approach to alleviat ing the preference elicitation bottleneck. As suming the existence of a population of users from whom we have elicited complete or in complete preference structures, we propose eliciting the preferences of a new user inter actively and incrementally, using the closest existing preference structures as potential de faults. Since a notion of closeness demands a measure of distance among preference struc tures, this paper takes the first step of study ing various distance measures over fully and partially specified preference structures. We explore the use of Euclidean distance, Spear man's footrule, and define a new measure, the probabilistic distance. We provide computa tional techniques for all three measures.