Summarizing Vocabularies in the Global Semantic Web

In the Semantic Web, vocabularies are defined andshared among knowledge workers to describe linked data for scientific,industrial or daily life usage. With the rapid growth of onlinevocabularies, there is an emergent need for approaches helping usersunderstand vocabularies quickly. In this paper, we study thesummarization of vocabularies to help users understand vocabularies.Vocabulary summarization is based on the structural analysis andpragmatics statistics in the global Semantic Web. Local BipartiteModel and Expanded Bipartite Model of a vocabulary are proposed tocharacterize the structure in a vocabulary and links betweenvocabularies. A structural importance for each RDF sentence in thevocabulary is assessed using link analysis. Meanwhile, pragmaticsimportance of each RDF sentence is assessed using the statistics ofinstantiation of its terms in the Semantic Web. Summaries are producedby extracting important RDF sentences in vocabularies under are-ranking strategy. Preliminary experiments show that it is feasibleto help users understand a vocabulary through its summary.