Case-Based Aggregation of Preferences for Group Recommenders

Abstract

We extend a group recommender system with a case base of previous group recommendation events. We show that this offers a new way of aggregating the predicted ratings of the group members. Using user-user similarity, we align individuals from the active group with indi viduals from the groups in the cases. Then, using item-item similarity, we transfer the preferences of the groups in the cases over to the group that is seeking a recommendation. The advantage of a case-based approach to preference aggregation is that it does not require us to commit to a model of social behaviour, expressed in a set of formulae, that may not be valid across all groups. Rather, the CBR system’s aggregation of the predicted ratings will be a lazy and local generalization of the behaviours captured by the neighbouring cases in the case base.