|
View:
New views
2 Messages
—
Rating Filter:
Alert me
|
|
|
Domain and Range of SubpropertiesDear Protégé
community,
I am
currently working on an ontology that integrates the Dublin Core and the
SWRC vocabulary. I would now like to specify these vocabularies for my purposes
by defining more specialized subproperties of the imported properties of Dublin
Core and SWRC.
I noticed that in
Protégé a subproperty always inherits the domain and range of the superproperty.
For
example, the property "dc:publisher" is defined with the domain "swrc:Document"
and the range "swrc:Organization" and "swrc:Person". If I create a supbproperty
of "dc:publisher" int the property tab, it also has the domain "swrc:Document"
and the range "swrc:Organization" and "swrc:Person".
Nevertheless, I
would like to define two subproperties of "dc:publisher" with more specific
domains, having one property with the domain "swrc:Organization" and another one
with the domain "swrc:Person".
If I create a
subproperty of "dc:publisher" the automatically inherited domain and range
classes cannot be deleted.
Nevertheless, I can
create a new property "voc:publisher_Person" with the Domain
"swrc:Document" and the range "swrc:Person" and then turn it into a subproperty
of "dc.publisher" via drag & drop in the property tab. Invoking the Pellet
reasoner for a consistency check does not lead to any error messages in this
case.
Is there a reason
why Protégé does not support this way of modelling subproperties whose domains
or ranges specify that of the superproperty? Maybe I am missing some modelling
paradigm that leads to inconsistency?
(I am working with
the Protégé 3.4 beta version.)
Thanks in advance
for your comments.
Carola
_______________________________________________ protege-discussion mailing list protege-discussion@... https://mailman.stanford.edu/mailman/listinfo/protege-discussion Instructions for unsubscribing: http://protege.stanford.edu/doc/faq.html#01a.03 |
|
|
Re: Domain and Range of SubpropertiesNo, there is no reason. I could actually not reproduce this exact
behavior. When I created the subproperty and added a new, more specific domain, the UI has replaced the generic domain with the more specific one. In any case, you should not be allowed to remove an inherited domain. Please ask in future protege-owl related questions on the protege-owl mailing list. This list is used only for protege-frames discussions. Thanks, Tania Carstens, Carola wrote: > Dear Protégé community, > > I am currently working on an ontology that integrates the Dublin Core > and the SWRC vocabulary. I would now like to specify these > vocabularies for my purposes by defining more specialized > subproperties of the imported properties of Dublin Core and SWRC. > > I noticed that in Protégé a subproperty always inherits the domain and > range of the superproperty. For example, the property "dc:publisher" > is defined with the domain "swrc:Document" and the range > "swrc:Organization" and "swrc:Person". If I create a supbproperty of > "dc:publisher" int the property tab, it also has the domain > "swrc:Document" and the range "swrc:Organization" and "swrc:Person". > > Nevertheless, I would like to define two subproperties of > "dc:publisher" with more specific domains, having one property with > the domain "swrc:Organization" and another one with the domain > "swrc:Person". > > If I create a subproperty of "dc:publisher" the automatically > inherited domain and range classes cannot be deleted. > Nevertheless, I can create a new property "voc:publisher_Person" with > the Domain "swrc:Document" and the range "swrc:Person" and then turn > it into a subproperty of "dc.publisher" via drag & drop in the > property tab. Invoking the Pellet reasoner for a consistency check > does not lead to any error messages in this case. > > Is there a reason why Protégé does not support this way of modelling > subproperties whose domains or ranges specify that of the > superproperty? Maybe I am missing some modelling paradigm that leads > to inconsistency? > > (I am working with the Protégé 3.4 beta version.) > > Thanks in advance for your comments. > > Carola > > > > > > > > > > > > > ------------------------------------------------------------------------ > > _______________________________________________ > protege-discussion mailing list > protege-discussion@... > https://mailman.stanford.edu/mailman/listinfo/protege-discussion > > Instructions for unsubscribing: http://protege.stanford.edu/doc/faq.html#01a.03 > _______________________________________________ protege-discussion mailing list protege-discussion@... https://mailman.stanford.edu/mailman/listinfo/protege-discussion Instructions for unsubscribing: http://protege.stanford.edu/doc/faq.html#01a.03 |
| Free Forum Powered by Nabble | Forum Help |