Triple Tagging: Toward Bridging Folksonomy and Semantic Web

Collaborative tagging is increasingly drawing attention as a promising source of taxonomic information; it is often referred to as folksonomy. A tag is information that is added to a resource (such as pictures, a web page, or blog entries); a tag can be represented as an RDF statement. However, no explicit study has been undertaken to bridge the gap between annotated tags and RDF statements. This paper proposes a model to represent tag information using RDF statements. Particularly, we give an incremental extension path from keyword tagging, which we call single-tagging, to relational tagging, which we call double-tagging, finally to subject-predicate-object tagging called triple-tagging, which (theoretically) corresponds to RDF statements. We build a new tagging system called LegoNote as the first triple-tagging system, and report some preliminary experiments. Our work will clarify the natural direction from the current tagging system and provide insights to realize a full Semantic Web vision.