Influence in Ratings-Based Recommender Systems: An Algorithm-Independent Approach

Recommender systems have been shown to help users nd items of interest from among a large pool of potentially interesting items. Inuenc e is a measure of the eect of a user on the recommendations from a recommender system. Inuence is a powerful tool for understanding the workings of a recommender system. Experiments show that users have widely varying degrees of inuence in ratings-based recommender systems. Proposed inuence measures have been algorithm-specic, which limits their generality and comparability. We propose an algorithm-independent denition of inuence that can be applied to any ratings-based recommender system. We show experimentally that inuence may be eectiv ely estimated using simple, inexpensive metrics.