Towards Social Semantic Suggestive Tagging.

Abstract. The organization of the knowledge on the web is increasingly becoming a social task performed by online communities whose members share a common interest in classifying different types of information for a later retrieval. Collaborative tagging systems allow people to organize a set of resources of interest through unconstrained annotations based on free keywords commonly named tags. Suggestive tagging techniques support users in this organization process and have shown to be helpful also in fostering a quick convergence to a shared tag vocabulary. In this paper, we propose a tag recommender which relies on the content analysis of the resource to be tagged, as well as on the personal and collective tagging history. The main contribution of this work is a model which combines semantic content analysis methods with existing suggestive tagging techniques. The expected benefit is the improvement of the user experience in social bookmarking systems, and more generally in collaborative tagging systems.