This work focusses on bridging between folksonomies, which provide social but mainly flat and unstructured metadata on web resources, and semantic web ontologies, which instead design structured, machine-processable knowledge spaces. The main purpose is to capture emerging semantics in social tagging systems and to overcome the gap between Semantic Web and Web 2.0, by preserving the complementary advantages of social and ontology-driven methods for describing, categorizing and processing web content. As a way to bridge this gap, we propose a method for linking tags from a folksonomy to concepts of an existing ontology, adopting a statistic approach. We have applied the proposed method to the data collected through the art portal Arsmeteo, relating them to the concepts of an OWL ontology of emotions. Intuitively, by our method we try to capture the latent emotional semantics of the tags. Some of the artworks in Arsmeteo could be visited in real exhibitions. In order to capture the emoti...