Leveraging semantic similarity for folksonomy-based recommendation

Abstract

For recommending interesting resources, such as Web pages or pictures available in social tagging systems, assessing their similarity with user profiles is crucial. Here, we analyze the role of semantic similarity to calculate the resemblance between user profiles and published resources in folksonomies. Experiments carried out with data from two social sites showed that associating semantics to tags results in more accurate similarities among elements in tagging systems and, consequently, enhances recommendations.