Hybrid Content and Tag-based Profiles for Recommendation in Collaborative Tagging Systems

Collaborative tagging systems have grown in popularity over the Web in the last years on account of their simplicity to categorize and retrieve content using open-ended tags. The increasing number of users providing information about themselves through social tagging activities caused the emergence of tag-based profiling approaches, which assume that users expose their preferences for certain contents through tag assignments. On the other hand, numerous content-based profiling techniques have been developed to address the problem of obtaining accurate models of user information preferences in order to assist users with information-related tasks such as Web browsing or searching. In this paper we propose a hybrid user profiling strategy that takes advantage of both content-based profiles describing long-term information interests that a recommender system can acquired along time and interests revealed through tagging activities, with the goal of enhancing the interaction of users with a collaborative tagging system. Experimental results of using hybrid profiles for tag recommendation are reported and possible applications of these profiles for obtaining personalized recommendations in collaborative tagging systems are discussed.