Search Mailing List Archives
[protege-owl] How to detect undefined references in rdf:resource
bunsz at gmx.de
Fri Jan 22 09:25:48 PST 2010
Thank you guys,
I had hoped the (pellet-)reasoner can classify these kind of missing
individual-definitions somehow out of the box. I'm limited to
protege-owl-api because I'm tied to Protege 3.4.1 using S(Q)WRL and the
builtIns as well. I wish I don't have to extend my already written
programm, because it will be costly to iterate over all individuals in
order to check wheter they have a type or not.
Am 22.01.2010 18:11, schrieb Timothy Redmond:
> Jaroslaw Slawinski wrote:
>> Maybe this is a solution for you (but need write a short program (with
>> Jena/owlapi/pellet/maybe rdflib (python))): check if the individual
>> has only the owl:Thing type. It does not guarantee that only such the
>> individuals you obtain but maybe it is enough for you. Moreover, you
>> can list all individuals from a model and test them if they have
> This is exactly what I would suggest (to find untyped resources) with
> a couple of amendments. You will probably need to use an rdf api
> because the OWL api will try to fit the ontology into OWL 2 DL and
> hence will infer types when it can. (The OWL api does have an rdf api
> included so this is still an option that way). Second I would check
> if the resource has no type. The owl:Thing type would indicate that
> it has already been typed as an individual.
>> Best regards,
>> On Fri, Jan 22, 2010 at 5:16 AM, <bunsz at gmx.de> wrote:
>>> On Jan 20, 2010, at 10:31 AM, Buns wrote:
>>>>> I wonder if it is possible to detect a reference to a non existent
>>>> This doesn't mean anything. You have a name therefore it exists. It
>>>> is possible that you don't know anything about it (e.g. its type) but
>>>> that might be another question.
>>> Ok, you're totally right. But can I detect, that I don`t know
>>> anything about it? The syntax "#res" means the individual is to be
>>> defined in this ontology and not in an imported one, but supposed
>>> there is no definition nowhere, isn't this kind of an inconsistency?
>>> What about this UntypedResource class mentioned by Thomas?
>>>>> If I have a property like<p1:hasWhatever rdf:resource="#res">
>>>>> and "res" doesn't exist ... can this be detected by the reasoner
>>>>> somehow, or am I fooled again by the open world?
>>> Jetzt kostenlos herunterladen: Internet Explorer 8 und Mozilla
>>> Firefox 3.5 -
>>> sicherer, schneller und einfacher!
>>> protege-owl mailing list
>>> protege-owl at lists.stanford.edu
>>> Instructions for unsubscribing:
>> protege-owl mailing list
>> protege-owl at lists.stanford.edu
>> Instructions for unsubscribing:
> protege-owl mailing list
> protege-owl at lists.stanford.edu
> Instructions for unsubscribing:
More information about the protege-owl