Approximation and relaxation of semantic web path queries

摘要: Abstract Given the heterogeneity of complex graph data on web, such as RDF linked data, it is likely that a user wishing to query will lack full knowledge structure and its irregularities. Hence, providing flexible querying capabilities assist users in formulating their information seeking requirements highly desirable. In this paper we undertake detailed theoretical investigation approximation, relaxation, combination, for purpose. The language adopt comprises conjunctions regular path queries, thus encompassing recent extensions SPARQL allow paths graphs using expressions (SPARQL 1.1). To add standard notions approximation based edit distance, well relaxation RDFS inference rules. We show how both these can be integrated into single framework provide incremental evaluation algorithms run polynomial time size returning answers ranked order ‘distance’ from original query. also combine first two disparate ‘flex’ operation simultaneously applies conjunct, even greater flexibility users, but still retaining complexity ability return order.