Abstract
RDF plays a prominent role in the vision of the Semantic Web and is used in widespread application such as FOAF, Dublin Core and RSS. Besides querying RDF it is also important to reason about RDF data. Therefore it is desirable to have a language which is capable of both reasoning and querying RDF. We propose the RDF query language rdflog, which is based on logic programming. This allows not only to combine querying and reasoning but also to transfer efficient evaluation techniques of logic programming to the evaluation of RDF query languages. In particular we investigate how tabling and magic set evaluation may be used in this context. The semantics of rdflog may be defined in terms of Herbrand models. We investigate whether the evaluation may be made more efficient by considering only non-redundant answers, called lean answers in the RDF context. Implementing the RDFS entailment rules is easily achieved with our approach by adding the RDFS axioms as rules to a query. In contrast to other rule based RDF query languages we also allow the creation of blank nodes in rule heads.

URL:
http://rewerse.net/publications/rewerse-publications.html#REWERSE-RP-2007-011

BibTeX:

@inproceedings{REWERSE-RP-2007-011,
	author = {Fran\c{c}ois Bry and Tim Furche and Clemens Ley and Benedikt Linse},
	title = {RDFLOG---Taming Existence: A Logic-based Query Language for RDF},
	booktitle = {Proceedings of Deduktionstreffen 2007, Koblenz, Germany (26th--27th March 2007)},
	year = {2007},
	organization = {GI-Fachgruppe Deduktionssysteme (DedSys)},
	url = {http://rewerse.net/publications/rewerse-publications.html#REWERSE-RP-2007-011}
}