incomplete answer ?

0 views
Skip to first unread message

laurent pierre

unread,
May 30, 2013, 6:58:35 AM5/30/13
to sta...@clarkparsia.com
Dear All,

I get puzzled with having not the expected results with both queries :

./stardog query "testMiscDB;reasoning=RL" "PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>
SELECT  ?x  ?exp ?expSum
WHERE { ?x :exp ?exp .
?x :expSum ?expSum }"

producing only

+---------------------------------+-------+--------+
|                x                |  exp  | expSum |
+---------------------------------+-------+--------+
| http://www.lolostar.org/prec#n0 | 0     | 0      |
| http://www.lolostar.org/prec#n1 | 1     | 1      |
+---------------------------------+-------+--------+

when all the items n0, ...., n4 should have a expSum property.

This error may follows from :

SELECT  ?y ?x  ?exp 
WHERE {               
?y a :Green }"

giving only n0 instead of n0, n1 and n4

and 

SELECT  ?y ?x  ?exp 
WHERE {               
?y a :BlueOrRed }"

giving nothing instead of n2 and n3

I join the turtle file

Cheers

Laurent


prec.ttl

Héctor Pérez-Urbina

unread,
May 30, 2013, 9:47:42 AM5/30/13
to stardog
Dear Laurent,

Thanks for your email. I can see that your ontology includes the following axiom:

(1) BlueOrRed equivalentClass Blue or Red.

As stated, axiom (1) goes beyond the expressive power of the three profiles of OWL; therefore, we simply ignore it (take a look at the log).

You could break it down into

(1.1) BlueOrRed subClassOf Blue or Red.
(1.2) Blue or Red subClassOf BlueOrRed.

Axiom (1.2) is a valid axiom as it is equivalent to

(1.2.1) Blue subClassOf BlueOrRed.
(1.2.2) Red subClassOf BlueOrRed.

However, axiom (1.1) again is beyond the profiles as it introduces non-determinism (if we know that an individual is an instance of BlueOrRed, is it an instance of Blue, Red, or both?).

If you absolutely need disjunction, you might use Pellet instead (depending on the size of your ontology).


Laurent


--
-- --
You received this message because you are subscribed to the C&P "Stardog" group.
To post to this group, send email to sta...@clarkparsia.com
To unsubscribe from this group, send email to
stardog+u...@clarkparsia.com
For more options, visit this group at
http://groups.google.com/a/clarkparsia.com/group/stardog?hl=en
 
 



--
Best,
Héctor

Laurent Pierre

unread,
May 30, 2013, 10:43:24 AM5/30/13
to sta...@clarkparsia.com
Dear Hector,

Thanks for your quick answer … I have to be more aware of the profile I use.  But still, if I get rid of axiom (1) an (1.1) and keep (1.2) (which seems sufficient for firing the rules), I don't get the expected results either.
Do I miss something again ?

Cheers

Laurent

Maurice Rabb

unread,
May 30, 2013, 11:47:58 AM5/30/13
to sta...@clarkparsia.com
Hi Héctor,

Is there an easy summary (of design pattern) available on how to compose disjunctive axioms?

Thanks,

Maurice

Héctor Pérez-Urbina

unread,
May 31, 2013, 5:01:28 PM5/31/13
to stardog
You are right. The inference should follow from 1.2.

The problem seems to be that some of your rules are cyclic. Currently, Stardog supports some cyclic rules by treating them as property paths. However, your rules can't be processed like this.

We are working on extending Stardog's capabilities in this regard. We'll keep you posted.

Héctor Pérez-Urbina

unread,
May 31, 2013, 5:02:46 PM5/31/13
to stardog
Hi Maurice,

I'm not sure what you mean.How to come up with them? How to use them? How to add them in an ontology?
Reply all
Reply to author
Forward
0 new messages