Applying the Technology Acceptance Model to Evaluation of Recommender Systems

MARCELO ARMENTANO; CHRISTENSEN, INGRID; SILVIA SCHIAFFINO

Polibits

Centro de Innovación y Desarrollo Tecnológico en Cómputo, Instituto Politécnico Nacional

Año: 2015 vol. 51 p. 73 - 79

2395-8618

In general, the study of recommender systemsemphasizes the efficiency of techniques to provide accuraterecommendations rather than factors influencing users?acceptance of the system; however, accuracy alone cannot accountfor users? satisfying experience. Bearing in mind this gap inthe research, we apply the technology acceptance model (TAM)to evaluate user acceptance of a recommender system in themovies domain. Within the basic TAM model, we incorporate anew latent variable representing self-assessed user skills to usea recommender system. The experiment included 116 users whoanswered a satisfaction survey after using a movie recommendersystem. The results evince that perceived usefulness of thesystem has more impact than perceived ease of use to motivateacceptance of recommendations. Additionally, users? previousskills strongly influence perceived ease of use, which directlyimpacts on perceived usefulness of the system. These findingscan assist developers of recommender systems in their attemptto maximize users? experience.