Hi Christian ,
sorry but there is no such functionality that is conveniently exposed.
Altough we do something very similar during the removal of statements to
figure out whether an inferred statement is still supported according to
some rule in the ruleset.
What you could do is to run a set of queries derived from each of the
rules in the ruleset.
If such a query return whatever solution, then the bindings will tell
you which are the exact premises that are present and match the
statement you are checking. Then you may dive into further to see what
else infers any of the premises (if not explicit).
for instance if you have a statement like (ex:Inst rdf:type
ex:SomeClass) and like to check whether the rdfs_2 rule support it (this
is the exact rule declaration from the pie file):
Id: rdfs2
x a y [Constraint a != ]
a z [Constraint z != ]
---
x z
then the query looks like that (a join between all the premises and the
consequent):
select * where {
?x ?a ?y . Filter(?a != rdf:type) .
?a rdfs:domain ?z. filter (?z != rdfs:Resource) .
# and add a check for the statement against the consequent
?x rdf:type ?z .
filter (?x = ex:Inst && ?z = ex:SomeClass) .
} limit 1
then if you get this result:
[?x = ex:Inst , ?a = ex:Prop , ?y = ex:something, ?z = ex:SomeClass]
it means that the statements from which it is inferred by this rule are:
[ex:Inst ex:Prop ?y = ex:something] and
[ex:Prop rdfs:Domain ex:SomeClass]
HTH,
Damyan Ognyanov.
Ontotext AD
On 12/15/2011 10:54 AM, grotheer wrote:
Hi everyone,
is there any way to find out information for triples generated by
reasoning, how the triple was derived?
I found out that Sesame currently has no functionality to do this.
Perhaps Big-/Swiftowlim can make this possible?
Thanks in advance,
Christian
PS: Sorry for my bad English.
___
OWLIM-discussion mailing list
OWLIM-discussion@ontotext.com
http://ontotext.com/mailman/listinfo/owlim-discussion
___
OWLIM-discussion mailing list
OWLIM-discussion@ontotext.com
http://ontotext.com/mailman/listinfo/owlim-discussion
