Ranking the Linked Data: The Case of DBpedia

摘要: The recent proliferation of crowd computing initiatives on the web calls for smarter methodologies and tools to annotate, query explore repositories. There is need scalable techniques able return also approximate results with respect a given as ranked set promising alternatives. In this paper we concentrate annotation retrieval software components, exploiting semantic tagging relying Linked Open Data. We focus DBpedia propose new hybrid methodology rank resources exploiting: (i) graphbased nature underlying RDF structure, (ii) context independent relations in graph (iii) external information sources such classical search engine social systems. compare our approach other similarity measures, proving validity algorithm an extensive evaluation involving real users.