The Impact of Recommender System and Users’ Behaviour on Choices’ Distribution and Quality

Abstract

Recommender Systems (RSs) research has become interested in assessing the system effect on the actual choices of their users: the distribution and the quality of the choices. By simulating users’ choices, influenced by RSs, it was shown that algorithmic biases, such as the tendency to recommend popular items, are transferred to the users’ choices. In this paper we conjecture that the effect of an RS on the quality and distribution of the users’ choices can also be influenced by the users’ tendency to prefer certain types of items, i.e., popular, recent, or highly rated items. To quantify this impact, we define alternative Choice Models (CMs) and we simulate their effect when users are exposed to recommendations. We find that RS biases can also be enforced by the CM, e.g., the tendency to concentrate the choices on a restricted number of items. Moreover, we discover that the quality of the choices can be jeopardised by a CM. We also find that for some RSs the impact of the CM is less prominent and their biases are not modified by the CM. This study show the importance of assessing algorithmic biases in conjunction with a proper model of users’ behaviour.