Resumo

VENCES-NAVA, Rodrigo; MENENDEZ-DOMINGUEZ, Víctor Hugo  e  MEDINA-PERALTA, Salvador. Evaluating a hybrid recommender system of theses. Ing. invest. y tecnol. [online]. 2019, vol.20, n.3, e025.  Epub 15-Out-2019. ISSN 2594-0732.  https://doi.org/10.22201/fi.25940732e.2019.20n3.025.

The task of evaluating a recommendation system is not limited to checking that the recommendation algorithm works correctly, but must also consider the validation that it fulfills the objectives for which it was designed. The latter varies depending on the recommendation system, since commercial systems increase sales, but for one of the education sector could be improving student achievement. This article focuses on the evaluation of a recommendation system, which presents to the user, degree jobs similar to those previously consulted based on different sources of information associated with them (metadata) and their interaction with the system and the community of users. Three types of studies were performed to evaluate the search engine's accuracy and retrieval, the use of keyword filtering in the recommendation and the latest utility, ease and perception of the system through the questionnaires System Usability Scale and Technology Acceptance Model. Statistical analyzes verify the effectiveness and acceptance of the Recommendation System by the three user groups in the experiments (teachers, theses’s students and students in general).

Palavras-chave : Recommender systems; SUS; TAM; evaluating recommender systems.

        · resumo em Espanhol     · texto em Espanhol     · Espanhol ( pdf )