Hello Barry,
 
No, ?text is not bound to a literal, it is an instance of one of my classes 
(nlp:Text).
As in my little testcase there is no ambiguity about that (nlp:hasLookupURI  
always has an instance of nlp:Text as object), I don't specify it in the query.
 
Thanks
Fabian

>>> Barry Norton <[email protected]> 21.03.2012 17:59 >>>

Hi, Fabian,

Is ?text bound to a literal when it succeeds in the two queries at the top?

Barry


On 20/03/2012 14:13, Fabian Cretton wrote: 


Hi,
 
I do have a strange behaviour that I don't understand, working with OWLIM 4.3.
 
I think there is no need to provide detailed information about the dataset to 
understand my problem, but will give more information if needed.
 
I have some texts with lookups on wordnet.
With a query, I want to find the texts which have a look-up that is 'linkedTo' 
a precise wordsense.
The linkedTo property being created by my self in the store.
 
If I execute 2 queries about two different wordsense, the results are correct: 
I get the same text as the answer
But if I merge the two wordsenses in a single query: the resultset is empty, 
and I can't understand why.
 
Query 1: the following query returns one result:
SELECT distinct ?text WHERE   {
?ann1 nlp:hasLookupURI ?ws1.
?ws1 wn20schema:linkedTo{1, 2} wn30:wordsense-prey-noun-2.
?ann1 nlp:isAnnAboutText ?text.
} 
 
Query 2: this other query returns the same result as query 1
SELECT distinct ?text WHERE   {
?ann2 nlp:hasLookupURI ?ws2.
?ws2 wn20schema:linkedTo{1, 2} wn30:wordsense-snake-noun-1.
?ann2 nlp:isAnnAboutText ?text.
} 
 
Query 3: now merging the two patterns should give me the same common result, 
but the resultset is empty:
SELECT distinct ?text WHERE   {
?ann1 nlp:hasLookupURI ?ws1.
?ws1 wn20schema:linkedTo{1, 2} wn30:wordsense-prey-noun-2.
?ann1 nlp:isAnnAboutText ?text.
?ann2 nlp:hasLookupURI ?ws2.
?ws2 wn20schema:linkedTo{1, 2} wn30:wordsense-snake-noun-1.
?ann2 nlp:isAnnAboutText ?text.
} 
 
Thank you for any help
Fabian


_______________________________________________
Owlim-discussion mailing 
[email protected]http://ontomail.semdata.org/cgi-bin/mailman/listinfo/owlim-discussion
_______________________________________________
Owlim-discussion mailing list
[email protected]
http://ontomail.semdata.org/cgi-bin/mailman/listinfo/owlim-discussion

Reply via email to