Re: Inferring Properties based on Types

On 10/1/07 12:16 PM, Michael Schneider wrote:
> Hi, Evren and Dmitry!
>
> @Dmitry: Yes, what you say below was also the first thing which came to my
> mind, when I read Evren's answer. Of course, I wanted to express that the
> property 'manman' is somehow "equivalent" to a given, *arbitrary* class
> 'Man'. The axiom set should of course not only allow models in which 'Man'
> is an equivalent class of 'owl:Thing'. :)
>   
Ah, of course :)
> But, Evren, many thanks for your reply! Great discussion, great points! I
> was completely unaware of this "Nonstructural Restrictions" issue, and I
> will look at it, as soon as I find time.
>   
When you read through that section you will actually see that those 
nonstructural restrictions do not apply to the set of axioms you had in 
the beginning :) It is ok to use the same property in (local or global) 
reflexivity restrictions and cardinality restrictions at the same time. 
The nonstructural restrictions only limit the usage of composite 
properties (the only composite property in your set of axioms is 
brother) so it wasn't a problem to begin with.

The lesson of this thread (at least for me) is not to answer any logic 
questions before noon on a Monday :)

Cheers,
Evren
> Hm, and perhaps, we will find another posibility? Or can it be shown that it
> is impossible?
>
> Let's hope and see...
>
> Cheers,
> Michael
>
>   
>> -----Original Message-----
>> From: Dmitry Tsarkov [mailto:dmitry.tsarkov@gmail.com] 
>> Sent: Monday, October 01, 2007 5:08 PM
>> To: Evren Sirin
>> Cc: Michael Schneider; Ian Horrocks; Swanson, Tim; Owl Dev
>> Subject: Re: Inferring Properties based on Types
>>
>> [reply to the list this time]
>>
>> On 10/1/07, Evren Sirin < evren@clarkparsia.com 
>> <mailto:evren@clarkparsia.com> > wrote:
>>
>>
>> [skip] 
>>
>>
>>
>>
>> 	With these modifications we get the
>> 	following set of axioms to express the above rule in OWL 1.1:
>> 	
>> 	    (R4) SubObjectPropertyOf( 
>> SubObjectPropertyChain(sibling manman) brother)
>> 	    (A1') ReflexiveObjectProperty( manman ) 
>> 	    (A3) ObjectPropertyDomain( manman Man )
>> 	    (A4) FunctionalObjectProperty( manman )
>>
>>
>> 	One could possibly define a macro that would translate 
>> a user-friendly
>> 	syntax into this set of axioms (and would generate the 
>> a unique property
>> 	like manman for different uses).
>>
>>
>> I think this wouldn't work as expected. To be more precise, it 
>> would silently introduce additional restrictions to the model. 
>> As the semantics of the construction 
>>          ReflexiveObjectProperty( manman )
>> is the following GCI 
>>          SubClassOf ( owl:Thing ObjectExistsSelf(manman) )
>> and we have a domain of the property manman, than the (implicit) axiom 
>>          SubClassOf (owl:Thing, Man)
>> would be added to the ontology. The same would happen for 
>> every other reflexive property with either range or domain 
>> defined. So, every concept in the ontology would be a Man (in 
>> addition to all other supertypes). 
>>
>> Best,
>> Dmitry.
>>
>>     
>
> --
> Dipl.-Inform. Michael Schneider
> FZI Forschungszentrum Informatik Karlsruhe
> Abtl. Information Process Engineering (IPE)
> Tel  : +49-721-9654-726
> Fax  : +49-721-9654-727
> Email: Michael.Schneider@fzi.de
> Web  : http://www.fzi.de/ipe/eng/mitarbeiter.php?id=555
>
> FZI Forschungszentrum Informatik an der Universität Karlsruhe
> Haid-und-Neu-Str. 10-14, D-76131 Karlsruhe
> Tel.: +49-721-9654-0, Fax: +49-721-9654-959
> Stiftung des bürgerlichen Rechts
> Az: 14-0563.1 Regierungspräsidium Karlsruhe
> Vorstand: Rüdiger Dillmann, Michael Flor, Jivka Ovtcharova, Rudi Studer
> Vorsitzender des Kuratoriums: Ministerialdirigent Günther Leßnerkraus
>
>
>

Received on Monday, 1 October 2007 20:00:08 UTC