Evaluation of User?s Engagement with a Movie Recommender System

MARCELO ARMENTANO; SILVIA SCHIAFFINO; ANALÍA AMANDI,

Foster City

Workshop; ACM RecSys Workshop on Recommender Systems Evaluation: Dimensions and Design (REDD 2014); 2014

ACM

In the eld of recommender systems, experimentation is usu- ally performed oine for assessing the accuracy of recom- mendations. However, to ensure a good user experience it is necessary to perform online experiments with real users to assess the factors that impact on the user's engagement with the system. In this article, we present results from an online experiment aimed at identifying dierent factors that could be considered to encourage the use of recommender systems. Our results show that the perceived usefulness of the system has more impact than the perceived ease of use in the user's engagement with the system and that the skills that the user believes he/she has to use recommender systems have a strong in uence on the perceived ease of use, which, in turn, has a strong in uence on the perceived usefulness. Further- more, the perceived usefulness partially mediates the impact of the simplicity of the graphical user interface on the user engagement with the system. Furthermore, we found that some of these relationships are dierent for low-experienced users than for high-experienced users.