Tech Report CS-09-07
Preference Aggregation in Group Recommender Systems for Committee Decision-Making
Jacob Baskin, Google and Shriram Krishnamurthi, Brown University
We present a preference aggregation algorithm designed for situations in which a limited number of users each review a small subset of a large (but finite) set of candidates. This algorithm aggregates scores by using users' relative preferences to search for a Kemeny-optimal ordering of items. We use variable-neighborhood local search, allowing the eachcient discovery of high-quality consensus orderings, which facilitates ective categorization of candidates while remaining computationally feasible for large problem instances. This algorithm provides a significant increase in solution quality over existing systems when evaluated against real-world data, as well as when tested against benchmark problem instances. We further discuss potential applications of this algorithm in group recommender systems for a variety of scenarios, including program committees and faculty searches, and discuss a number of considerations regarding its use.
(complete text in pdf)