Recommending Smart Tags in a Social Bookmarking System

Collaborative tagging systems are harnessing the power of online communities, making the task of knowledge contribution more attractive to a broader audience of Web users. In particular, social bookmarking systems have shifted the organization of bookmarks from an individual activity performed on a personal desktop to a collective endeavor over the Web. In such a context, suggestive tagging has proved to be helpful in consolidating the usage of tags, leading to a quick convergence to a folksonomy. In a social bookmarking system, users' annotations can be regarded as a reliable indicator of interests and preferences. A recommender system is able to learn user interests and preferences during the interaction in order to construct a user profile. In this paper, we propose a smart tag recommender able to learn from past user interaction as well as the content of the resources to annotate. The aim of the system is to support users of current social bookmarking systems by providing a list of new meaningful tags. The proposed system is based on ITem Recommender, a content-based recommender previously used in a Digital