Abstract

Abstract. While tags in collaborative tagging systems serve primarily an indexingpurpose, facilitating search and navigation of resources, the use of the same tags bymore than one individual can yield a collective classification schema. We present anapproach for making explicit the semantics behind the tag space in social taggingsystems, so that this collaborative organization can emerge in the form of groups ofconcepts and partial ontologies. This is achieved by using a combination of shallowpre-processing strategies and statistical techniques together with knowledge providedby ontologies available on the semantic web. Preliminary results on the del.icio.us andFlickr tag sets show that the approach is very promising: it generates clusters withhighly related tags corresponding to concepts in ontologies and meaningfulrelationships among subsets of these tags can be identified.

Links and resources

Tags

community