Authors: Silvia Stefanova, Tore Risch
Addresses: Department of Information Technology, Uppsala University, Box 337, Uppsala 75105, Sweden. ' Department of Information Technology, Uppsala University, Box 337, Uppsala 75105, Sweden
Abstract: Both Topic Maps and RDF are popular semantic web standards designed for machine processing of web documents. Since these representations were originally created for different purposes, they have differences both in their concepts and in their data models. The Topic Map data model can be seen as an ontology for building indices over web documents, whereas RDF is a language to define arbitrary properties of web resources. Ontologies can be defined using the RDF vocabulary description language RDF-Schema (RDFS) language. RDF(S) repositories can be searched using the query language SPARQL. To make Topic Maps exposed to RDF-based tools and searchable with SPARQL, our approach is to map Topic Maps to a view expressed in RDFS. The view can be queried using SPARQL. The problem concerning efficient query processing of SPARQL queries to the rather complex RDF view of Topic Maps has been studied in detail. Our approach has been applied on searching indices to e-government services.
Keywords: topic maps; RDFS repositories; RDF-Schema; views; query processing; SPARQL; semantic web standards; data models; ontology; indices; web documents; web resources; e-government services; electronic government; resource description framework.
International Journal of Metadata, Semantics and Ontologies, 2010 Vol.5 No.1, pp.1 - 16
Available online: 09 Apr 2010 *Full-text access for editors Access for subscribers Purchase this article Comment on this article