Recomendação de recursos educacionais para grupos em redes sociais

Abstract

With the growth of the amount of educational resources available on the web and as repositories usually include a wide range of subjects, including different educational goals, choosing an educational resource is still a challenge. The recommender systems represent a tool capable of contributing to the location and selection of these resources. However, to a recommender system generate satisfactory recommendations, it is important to identify information that helps to define the user’s profile. Lots of information about profile, interests and preferences of the users can be identified as a result of interactions that occur in social networks. Many times those users, besides organizing themselves into interest groups, during their time spent on social networks, seek for educational resources of their individual or collective interests. This work aims to explore the information generated by a users group on social networks in the search and selection process of educational resources, considering their members common interest. The proposal and the development of an architecture named BROAD-GRS are presented, able to set educational profile of Facebook social network users groups and make recommendations of educational resources to the group, respecting the members preferences and maximizing the general satisfaction of all users in the choice of recommended items. The recommendation is based on three sources of educational resources: repository of learning objects (BROAD), linked data repositories (DBpedia and Open University), and video repository (Youtube). The evaluation of the proposal was made with the development of a prototype and a case study and the results pointed out to the viability of the proposed solution.