Journal Ankesh

From Semantic Portal Wiki

Presentation given at CSCI 6966 Advanced Semantic Web (Fall 2008)#Lesson 10

• Speaker: Ankesh Khandelwal
• Title: Completeness, decidability and complexity of entailment for RDF Schema and a semantic extension involving the OWL vocabulary
• Authors: Herman J. ter Horst
• Conference: Journal of Web Semantics, 2005
• URL: http://www.websemanticsjournal.org/papers/20050719/document5.pdf
• Date of Presentation: 2008/11/06

Questions

ID	Question	Name	Answer
Journal Ankesh GTW 1	Section 1.5 first points out the authors discovery that the RDFS entailment rules are incomplete, demonstrated in that section with the example { p rdfs:subPropertyOf b . v p w . } failing to infer { v b w } because the blank node b cannot syntactically appear in the predicate position. The paper goes on to consider "generalized RDF graphs" which relax this syntactic restriction, allowing a completion of the RDFS entailment rules. Is this approach preferable to one in which RDFS simply disallows blank nodes to refer to properties? Can you give a (hopefully compelling) example where the relaxation is preferable to the restriction?	Gregory Todd Williams
Journal Ankesh Jesse Weaver	Great article; I'm glad to have had the chance to read it in this course. The findings of this journal article seem significant, particularly the differing decidable entailments that are NP, or P in the absence of blank nodes. Oracle has based their OWLSIF entailment on pD* (see http://www.w3.org/2007/OWL/wiki/OracleOwlPrime). Are you aware of any other major impacts this article has had in industry? or any other major impacts in academia? Also, (just out of curiosity) the authors mention that in pD*, there is no restriction on treating classes and properties as instances. While this is perhaps ideal in the logic world (and is also supported in OWL Full), are you aware of any great demand for this feature?	Jesse Weaver

Attendees

Tim Lebo