W3C home > Mailing lists > Public > public-owl-dev@w3.org > January to March 2012

Re: SILK context and thanks for replies to my question on translators from OWL 2 DL/Full to FOL

From: Michael Schneider <schneid@fzi.de>
Date: Tue, 17 Jan 2012 13:36:33 +0100
Message-ID: <4F156B51.1080703@fzi.de>
To: Benjamin Grosof <BenjaminG@vulcan.com>
CC: "Bijan Parsia (bparsia@cs.man.ac.uk)" <bparsia@cs.man.ac.uk>, 'Chris Mungall' <cjmungall@lbl.gov>, 'Mike Dean' <mdean@bbn.com>, "Michael Kifer (kifer@cs.sunysb.edu)" <kifer@cs.sunysb.edu>, "'public-owl-dev@w3.org'" <public-owl-dev@w3.org>
Hi Benjamin!

Just a clarification:

Am 17.01.2012 01:29, schrieb Benjamin Grosof:

> The mapping from OWL to SILK is designed mainly in
> terms of FOL and the best way to implement it is via FOL if not directly
> as a visitor on the OWL 2 API or something similar.

When Bijan proposes to use an OWL API Visitor (and that's what I am 
doing myself), he probably means that this is a rather controlled 
approach to get from the different serialization formats of OWL to a FOL 
representation:

   1) Use one of the OWL API parsers to parse RDF/XML, Functional 
Syntax, Manchester Syntax, or whatever OWL syntax into an OWLOntology 
object, which is a representation closely following the OWL 2 Structural 
Specification, i.e., the abstract syntax of OWL 2 (DL).

   2) There are different "Visitor" interfaces for OWL 2 axioms, class 
expressions and other language constructs, which are applied to the 
different axioms of an OWLOntology object to do something with them. In 
your case, you would implement a visitor that maps OWL 2 axioms to their 
respective FOL representation.

So the result will indeed be an FOL representation of the input ontology 
w.r.t. the OWL 2 Direct Semantics, and after the translation process, 
there would be no dependency on the OWL API anymore, so the OWL API is 
just used as a vehicle. The actual translation would still need to be 
defined by yourself within such a visitor. But you would get the 
translation from the different OWL serializations basically for free 
(thanks to the OWL API), and by using the visitors, the translation job 
is well-guided, since the visitor interfaces know about all the OWL 
language features, so you simply need to implement the respective 
methods: you can fully focus on the actual FOL translation! And if you 
closely follow the definitions given in the OWL 2 Direct Semantics, 
which essentially /is/ a mapping from the OWL 2 abstract syntax to an 
FOL/set-theory mix (it's all rewritable in FOL, afterall), even this is 
mainly a no-brainer, which just takes a couple of days. :-)

Of course, there will be other ways to do it, but, for me at least, it's 
the most straight-forward and convenient way to implement a FOL 
translation for OWL 2 DL (plus a bit).

Cheers,
Michael

-- 
.........................................................
Dipl.-Inform. Michael Schneider
Research Scientist, IPE / WIM

FZI Forschungszentrum Informatik
Haid-und-Neu-Str. 10–14
76131 Karlsruhe, Germany
Tel.: +49 721 9654-726
Fax: +49 721 9654-727

michael.schneider@fzi.de
www.fzi.de

.........................................................
Forschungszentrum Informatik (FZI) an der Universität Karlsruhe
Stiftung des bürgerlichen Rechts
Stiftung Az: 14-0563.1 Regierungspräsidium Karlsruhe
Vorstand: Dipl. Wi.-Ing. Michael Flor, Prof. Dr. Ralf Reussner,
Prof. Dr. Rudi Studer, Prof. Dr.-Ing. J. Marius Zöllner
Vorsitzender des Kuratoriums: Ministerialdirigent Günther Leßnerkraus
.........................................................
Received on Tuesday, 17 January 2012 12:36:58 GMT

This archive was generated by hypermail 2.3.1 : Wednesday, 27 March 2013 09:32:59 GMT