Hello Fabian,
It seems that your latest example is another reproduction of the same
problem in OWLIM-Lite. We will add this to the bug tracker and look to
release a fix in the near future.
Regarding your question:
is that right that SPARQL property path with "+" don't need the OWL
property to be "transitive"?
The answer is 'NO', SPARQL is completely independent of the entailment
regime and so it doesn't matter what 'kinds' of properties link resources.
I hope this helps,
barry
--
Barry Bishop
OWLIM Product Manager
Ontotext AD
Tel: +43 650 2000 237
email: [email protected]
www.ontotext.com
On 22/12/11 11:24, Fabian Cretton wrote:
Hello Ivan,
It might be a bug, or it could just be my wrong understanding of
property paths.
Here is something similar where you could tell me if I am wrong, and
if not, could illustrate the same problem.
First question: is that right that SPARQL property path with "+" don't
need the OWL property to be "transitive" ? from the following test
that I do, it doesn't need to be transitive, and here I describe what
is for me a strange behaviour:
- I work with the W3C version of wordnet 2.0
- the T-Box "wnfull.rdfs" was edited, so that hyponymOf is no more
transitive
- I load the rdf file in a OWLIM Lite 4.3 repository, with a
"owl-horst-optimized" ruleset
As the property hyponymOf is not transitive in my edited file, the
following query:
PREFIX wn20schema: <http://www.w3.org/2006/03/wn/wn20/schema/>
PREFIX wn20instances: <http://www.w3.org/2006/03/wn/wn20/instances/>
select * WHERE {wn20instances:synset-rat-noun-1 wn20schema:hyponymOf
?hypo}
will give the asserted hyponym of rat: wn20instances:synset-rodent-noun-1
The following query gives me all the hyponyms of rat:
select * WHERE {wn20instances:synset-rat-noun-1 wn20schema:hyponymOf+
?hypo}
The result being (I put them in the 'asserted' order):
wn20instances:synset-rodent-noun-1
wn20instances:synset-placental-noun-1
wn20instances:synset-mammal-noun-1
wn20instances:synset-vertebrate-noun-1
wn20instances:synset-chordate-noun-1
wn20instances:synset-animal-noun-1
so rat is hyponymOf animal
But the following query doesn't return any result:
select * WHERE {
wn20instances:synset-rat-noun-1 wn20schema:hyponymOf+ ?hypo.
FILTER(?hypo = wn20instances:synset-animal-noun-1)
}
is it normal ?
and so the following 'ask' returns 'NO' -> where I would expect 'YES':
ask WHERE {
wn20instances:synset-rat-noun-1 wn20schema:hyponymOf+
wn20instances:synset-animal-noun-1.
}
Thanks for any help
Fabian
>>> Ivan Peikov <[email protected]> 22.12.2011 10:25 >>>
Hi Fabian,
The two queries are indeed functionally equivalent so as far as I can
tell
this is a bug in OWLIM-Lite.
As a temporary workaround I can only advise you to use the working
form (#2).
I will file the issue and we'll address it as soon as possible. In the
mean
time can you possibly send the data (or the relevant part of it) that
could
help us reproduce it locally?
Thanks!
Cheers,
Ivan
On Thursday 22 December 2011 10:53:03 Fabian Cretton wrote:
> Thank you
> Fabian
>
> >>> Barry Bishop <[email protected]> 22.12.2011 09:42 >>>
>
> Hi Fabian,
>
> I'm not sure about this, so I'll pass it over to the developers. They
> should get back to you with an answer shortly.
>
> Regards,
> barry
>
> --
> Barry Bishop
> OWLIM Product Manager
> Ontotext AD
> Tel: +43 650 2000 237
> email: [email protected]
> On 19/12/11 14:56, Fabian Cretton wrote:
>
> Dear all,
>
> Is there a difference between the two following patterns:
> ?ws1 wn20schema:inSynset ?syn1.
> ?ws2 wn20schema:inSynset ?syn2.
> ?syn1 wn20schema:hyponymOf+ ?syn2.
>
> compared to:
> ?ws1 wn20schema:inSynset ?syn1.
> ?ws2 wn20schema:inSynset ?syn2.
> ?syn1 wn20schema:hyponymOf+ ?tst.
> FILTER(?syn2 = ?tst)
>
> I use this in a larger queries, where the first form doesn't return the
> results I would expect, but the second one does, is this normal ? (I
could
> post information about the data and the full query if needed).
>
> I am working with OWLIM-Lite 4.3
>
> Thanks for any help
> Fabian
> _______________________________________________
> OWLIM-discussion mailing
>
[email protected]http://ontotext.com/mailman/listinfo/owli
> m-discussion
_______________________________________________
OWLIM-discussion mailing list
[email protected]
http://ontotext.com/mailman/listinfo/owlim-discussion
_______________________________________________
OWLIM-discussion mailing list
[email protected]
http://ontotext.com/mailman/listinfo/owlim-discussion
_______________________________________________
OWLIM-discussion mailing list
[email protected]
http://ontotext.com/mailman/listinfo/owlim-discussion
