John; The need to set "Complete Mode" is an artifact of one of the few
places where Composer's UI automatically does entailments. Since the
UI will recognizes that members-of-a-subclass-are-members-of-the-
class, then there isn't a reason to materialize the inferences in the
Inferences tab (I assume that is what you mean here). If you were to
run this via a service - a SPARQLMotion script for example - you will
get all of the entailments by default.
I find the example a bit confusing. I take it you mean a SPIN rule.
Any SPIN rule that doesn't use ?this is suspect, and that is the case
here. The way SPIN works is that it executes the rule against all
members of a class (and its subclasses) by binding ?this to the rule
and executing the query. This localizes the query so it is only
executed against the instance.
Given that you place the query in the spin:rules property of
publishedBooks:Book. You can remove the triple pattern {?book a
publishedBooks:Book}, which is superfluous because the rule will
automatically only apply to members of publishedBooks:Book.
But this is where I get confused with your example, because it seems
to just copy triples to itself - a null op, basically. So let me try
a different example, if just to show how ?this works. Assume that you
have members of foaf:Person. The following rule would infer that all
book authors are members of the Author class:
CONSTRUCT
{ ?author a publishedBooks:Author
}
WHERE
{ ?this dc:creator ?author .
}
So I take it you have configured inferencing to run both OWLIM and
TopSPIN? You can avoid that and run just one reasoner by selecting
the OWL 2 RL profile in Composer. Go tot he Ontology home and select
the Profile sub-tab (see Help > User Interface Overview > Resource
Editor > Ontology Profile) and choose OWL 2 RL from the library. This
will run OWL inferences and you can configure inferences to just run
TopSPIN and it will take care of both your customized rules and OWL
entailments.
-- Scott
On Sep 20, 4:02 pm, John Perdoni <[email protected]> wrote:
> Hi again,
>
> I thought I had a reasonable grasp on inferencing, but back to newbie
> status lol.
>
> I had a classhttp://example.com/library/publishedBooks#Bookwith 2
> instances which i ran the following script on
>
> CONSTRUCT {
> ?book a publishedBooks:Book .
> ?book ?p ?o .}
>
> WHERE {
> ?book a publishedBooks:Book .
> ?book ?p ?o .
>
> }
>
> All good up until now
>
> Thinking about the future I decided to add a subclass/es to try and
> classify the type of book
> iehttp://example.com/library/publishedBooks#Electrical which is a
> rdfs:subClassOf example.com:Book and moved the instances to this subClass
>
> This is where my so called knowledge (or lack of it) gets blown out of the
> window as running the same rule now provides no results.
>
> I was under the impression that when running owl-max inferencing with
> TopSpin(Sparql rules) on the above two instances would infer the following
>
> ie CONSTRUCT {?instance rdf:type ?book}
> WHERE{?sub rdfs:subclassOf ?book.
> ?instance rdf:type ?sub}
>
> and would result in
>
> ?instance a ?Book
>
> I can only achieve this result if I i tick the compleMode check box under
> inferencing configuration, but this is not something I thought I should do.
>
> Stangely this has never been an issue in the past and is not an issue if i
> run owl-max inferencing with the same data on a Sesame repository
>
> Sorry if this yet again is lack of knowledge
>
> John
--
-- You received this message because you are subscribed to the Google
Group "TopBraid Suite Users", the topics of which include Enterprise Vocabulary
Network (EVN), TopBraid Composer, TopBraid Live,
TopBraid Ensemble, SPARQLMotion, SPARQL Web Pages and SPIN.
To post to this group, send email to
[email protected]
To unsubscribe from this group, send email to
[email protected]
For more options, visit this group at
http://groups.google.com/group/topbraid-users?hl=en
