Preliminary Results of Logical Ontology Pattern Detection using SPARQL and Lexical Heuristics

Ondrej Sváb-Zamazal, François Scharffe, Vojtech Svátek
2009 International Semantic Web Conference  
Ontology design patterns were proposed in order to assist the ontology engineering task, providing models of specific construction representing a particular form of knowledge. Various kinds of patterns have since been introduced and classes of patterns identified. Detecting these patterns in existing ontologies is needed in various scenarios, for example the detection of the the two parts of an alignment pattern in an ontology matching scenario, or the detection of an anti-pattern in an
more » ... tion scenario. In this paper we present a novel method for the detection of logical patterns in ontologies. This method is based on both SPARQL, as the underlying language for retrieving patterns, and a lexical heuristic constraining the query. It extends our previous works on ontology patterns modeling and detection. We describe an algorithm computing a tokenbased similarity measure used as the lexical heuristic. We conduct an experiment on a large number of Web ontologies, obtaining interesting measures on the usage frequency of three selected patterns.
dblp:conf/semweb/Svab-ZamazalSS09 fatcat:gnpkgtbv7jbajiitvrxwaf6mwi