作者: Deborah L. McGuinness , Tim Finin , Paulo Pinheiro da Silva , Anupam Joshi , Yun Peng
DOI: 10.13016/M2J38KN3P
关键词: Information retrieval 、 RDF Schema 、 Computer science 、 RDF 、 Blank node 、 Swoogle 、 RDF/XML 、 Cwm 、 Linked data 、 SPARQL
摘要: The Semantic Web facilitates integrating partial knowledge and finding evidence for hypothesis from web sources. However, the appropriate level of granularity tracking provenance RDF graph remains in debate. document is too coarse since it could contain irrelevant information. triple will fail when two triples share same blank node. Therefore, this paper investigates lossless decomposition using molecule, which finest component an graph. A sub-graph {\em lossless} if can be used to restore original without introducing new triples. finest} cannot further decomposed into sub-graphs. algorithms molecule have been formalized implemented by a prototype service Swoogle project.