A field study of a visual controllable talk recommender. In Proceedings of the 2013 Chilean Conference on Human – Computer Interaction

Abstract

In this paper, we contribute to the study of recommender systems from a HCI perspective by investigating the effects upon the user experience of a novel interface which uses a Venn diagram to represent the outputs of an interactive talk recommender system. We present the results of a preliminary user study on talk recommendations in the context of a conference with n=37 people that used our system under one of two conditions: a static list of recommendations, or the enhanced visual controllable interface. The user behavioral analysis and the results of a survey that n=17 users answered provide interesting insights for designers and developers of interfaces for recommender systems, especially when the items can have one or more contexts of relevancy as in a hybrid recommender system.