W3C home > Mailing lists > Public > public-owl-dev@w3.org > April to June 2008

Re: Intersection of properties

From: Uli Sattler <sattler@cs.man.ac.uk>
Date: Mon, 7 Apr 2008 11:21:28 +0100
Cc: <public-owl-dev@w3.org>
Message-Id: <37F8BDF6-830C-4076-BF81-D53A85F35ADB@cs.man.ac.uk>
To: "Dimitrios Koutsomitropoulos" <kotsomit@hpclab.ceid.upatras.gr>


On 6 Apr 2008, at 17:34, Dimitrios Koutsomitropoulos wrote:
>
> Hello,
>
> Given that OWL 1.1 allows for disjoint properties, how difficult  
> would it be
> to implement properties intersection

not too difficult as long as it is intersection of simple properties  
-- but i don't have a pointer

> and/or complement

more tricky since you loose "locality": if you say that Mary is an  
instance of

allValuesFrom ComplementOf(kissed) ComplementOf(Boys),

then you have said that Mary kissed all (!) Boys, and there can be  
quite many of them. For this, look e.g. at

C. Lutz and U. Sattler. Mary likes all Cats. In F. Baader and U.  
Sattler, editors, Proceedings of the 2000 International Workshop in  
Description Logics (DL2000), number 33 in CEUR-WS, pages 213-226,  
Aachen, Germany, August 2000. RWTH Aachen. Proceedings online  
available from http://SunSITE.Informatik.RWTH-Aachen.DE/Publications/CEUR-WS/Vol-33/ 
.

It does not lead to undecidability, but requires different  
reasoning...see for example

Using Tableau to Decide Expressive Description Logics with Role  
Negation. Together with D. Tishkovsky. In Aberer, K. et al (eds), The  
Semantic Web, 6th International Semantic Web Conference, 2nd Asian  
Semantic Web Conference, ISWC 2007 + ASWC 2007, Busan, Korea, November  
11-15, 2007. Lecture Notes in Computer Science, Vol. 4825, Springer,  
438-451.

Cheers, Uli

> ? Would this lead to
> undecidability? If so, could you please provide appropriate  
> properties or
> explain?
>
>
> Regards,
>
> Dimitrios
>
>
Received on Monday, 7 April 2008 10:32:50 GMT

This archive was generated by hypermail 2.3.1 : Wednesday, 27 March 2013 09:32:56 GMT