The current industry standard for recommender system uses variants of collaborative filtering (CF), where recipient-source relationships are determined by the extent to which the recipient and source share interests. This research attempts to improve the performance of these CF recommender systems by identifying additional measures of relationship indicators based on theories from communication and marketing. We developed a social filtering model that incorporates these various social measures (e.g. Trust, Reputation, Interaction Frequency, and Relationship Duration), and conducted an empirical study to test the model. The results from the study show small, but significant, improvements for various social relationships. We plan to build on these preliminary results to further consolidate our research on using social relationship in recommender systems.
|State||Published - 1 Jan 2007|
|Event||17th Workshop on Information Technologies and Systems, WITS 2007 - Montreal, QC, Canada|
Duration: 8 Dec 2007 → 9 Dec 2007
|Conference||17th Workshop on Information Technologies and Systems, WITS 2007|
|Period||8/12/07 → 9/12/07|