A comparison of calibrated and intent-aware recommendations

Abstract

Calibrated and intent-aware recommendation are recent approaches to recommendation that have apparent similarities. Both try, to a certain extent, to cover the user's interests, as revealed by her user profle. In this paper, we compare them in detail. On two datasets, we show the extent to which intent-aware recommendations are calibrated and the extent to which calibrated recommendations are diverse. We consider two ways of defning a user's interests, one based on item features, the other based on subprofles of the user's profle. We fnd that defning interests in terms of subprofles results in highest precision and the best relevance/diversity trade-of. Along the way, we defne a new version of calibrated recommendation and three new evaluation metrics.