Enhancing Recommender Systems with Access to Electronic Health Records and Groups of Interest in Social Networks

Recommender systems have been around for several years as tools intended to discover items that are likely to be of interest for the users, as inferred from profiles that gather information about web pages they visit, TV programs they watch or commercial products they purchase. In this paper, we introduce health-related aspects as a new dimension of the user profiles, enabling a number of features that may have a significant impact in the field of interactive services to the home. Specifically, we have enhanced an existing recommender system to process information stored in electronic health records and in groups of interest created within social networks. This has made it necessary to refine the recommendation logic in order to reckon the enabling/disabling nature of health-related data, and also to deal with the risks that arise from ignorance or commercial interest in the social context.