TY - JOUR
T1 - SOWL QL
T2 - Querying Spatio-Temporal Ontologies in OWL
AU - Stravoskoufos, Konstantinos
AU - Petrakis, Euripides G M
AU - Mainas, Nikolaos
AU - Batsakis, Sotirios
AU - Samoladas, Vasilis
PY - 2016/12
Y1 - 2016/12
N2 - We introduce SOWL QL, a query language for spatio-temporal information in ontologies. Building-upon SOWL (Spatio-Temporal OWL), an ontology for handling spatio-temporal information in OWL, SOWL QL supports querying over qualitative spatio-temporal information (expressed using natural language expressions such as “before”, “after”, “north of”, “south of”) rather than merely quantitative information (exact dates, times, locations). SOWL QL extends SPARQL with a powerful set of temporal and spatial operators, including temporal Allen topological, spatial directional and topological operations or combinations of the above. SOWL QL maintains simplicity of expression, and also upward and downward compatibility with SPARQL. Query translation in SOWL QL yields SPARQL queries, implying that querying spatio-temporal ontologies using SPARQL is still feasible but suffers from several drawbacks, the most important of them being that, queries in SPARQL become particularly complicated and users must be familiar with the underlying spatio-temporal representation (the “N-ary relations” or the “4D-fluents” approach in this work). Finally, querying in SOWL QL is supported by the SOWL reasoner which is not part of the standard SPARQL translation. The run-time performance of SOWL QL has been assessed experimentally in a real data setting. A critical analysis of its performance is also presented.
AB - We introduce SOWL QL, a query language for spatio-temporal information in ontologies. Building-upon SOWL (Spatio-Temporal OWL), an ontology for handling spatio-temporal information in OWL, SOWL QL supports querying over qualitative spatio-temporal information (expressed using natural language expressions such as “before”, “after”, “north of”, “south of”) rather than merely quantitative information (exact dates, times, locations). SOWL QL extends SPARQL with a powerful set of temporal and spatial operators, including temporal Allen topological, spatial directional and topological operations or combinations of the above. SOWL QL maintains simplicity of expression, and also upward and downward compatibility with SPARQL. Query translation in SOWL QL yields SPARQL queries, implying that querying spatio-temporal ontologies using SPARQL is still feasible but suffers from several drawbacks, the most important of them being that, queries in SPARQL become particularly complicated and users must be familiar with the underlying spatio-temporal representation (the “N-ary relations” or the “4D-fluents” approach in this work). Finally, querying in SOWL QL is supported by the SOWL reasoner which is not part of the standard SPARQL translation. The run-time performance of SOWL QL has been assessed experimentally in a real data setting. A critical analysis of its performance is also presented.
KW - Query language
KW - Spatio-temporal ontology
UR - http://www.scopus.com/inward/record.url?scp=84992343513&partnerID=8YFLogxK
U2 - 10.1007/s13740-016-0064-5
DO - 10.1007/s13740-016-0064-5
M3 - Article
AN - SCOPUS:84992343513
VL - 5
SP - 249
EP - 269
JO - Journal on Data Semantics
JF - Journal on Data Semantics
SN - 1861-2032
IS - 4
ER -