On 9/21/11 3:05 PM, Bohms, H.M. (Michel) wrote:
- wrt locked: indeed
in "Topbraid", I'll make a new project (strange that in the
past I never had this problem...maybe sometime you started
to lock stuff there...?
- wrt closed world:
for references I know that having card constraints , say
card=1 and two refs then actually inferred is ref1 = ref2
...but how does this work with values..like with the
areas....you can hardly infer that 23 =42 .....
Still not OWL, but you can easily write a SPIN rule that
would do this. OWL equivalence is done with owl:sameAs, so you
could define {:ref1 owl:sameAs :ref2}. Since owl:sameAs is an
object property relationship, you won't get into the bind of
defining equivalence of literals.
In terms of rules for cardinality, see the specification of OWL 2
RL at
http://www.w3.org/TR/owl2-profiles/#Reasoning_in_OWL_2_RL_and_RDF_Graphs_using_Rules.
The cls-max set of rules define all that OWL can reasonably do
with cardinality restrictions.
-- Scott
- anyway I will use
the cw-variants....
thanks for patience
and help, Michel
Michel, see responses inline:
On 9/21/11 1:37 PM, Bohms, H.M. (Michel) wrote:
1. Yes, just before I sent you I cleaned up a bit taking the instances out. When I create 1 instance called Rectangle_1 and do different sets of inputs (and sevral infers) I get multiple values for area. "refersh and show problems' does not flag.
2. But ok, I did not explcitly define my card constraints like your aextra ask query below! I just used normal class axioms, here like "area
exactly one"
The semantics you are looking for are not supported by OWL.
The Open World Assumption (OWA) prevent this kind of
inference. You can look that up in a variety of places. And
it is a common misconception about OWL inferences and
consistency checking (please note that OWL includes some
consistency checks, but is NOT a constraint checking
language).
You can, however use the CWA (Closed World Assumption) axioms
in the profile. Just as you did with OWL 2 RL, go to the
Ontology Profile tab and select "SPIN Constraints on OWL
axioms with CWA". The cardinality checks are included in
these rules.
So is it true that those constraints (restrictions) will not be flagged a t all (only the explcitly defined ones)?
In that case its clear....
Try the CWA rules. I think this has
the semantics you are looking for.
3. currently I do: non-incremental/iteratily
But....wrt 2. I guess it doesn't matter much what I say here...(?) (indeed when I specify incremental>no difference, still multiple values and no constraint violation warning...
OK, now I'm able to infer what you are doing. Keep in mind
that OWL is a monotonic logic. You can add facts, but you
can't remove them. So if you run inferences, change a value,
then run inferences again, the result is two values. This is
precisely how OWL is specified.
If you need to modify data, then you have to look outside of
OWL, as it cannot support removing old values. Facts cannot
be removed. SPARQL is another W3C semantic web technology,
and SPARQL Update supports delete and modify (delete/insert
combo).
Incremental inferencing is a specialized TBC-ME feature that
supports re-inferencing after a value has changed. Basically
it analyzes rules that specify the modified value, removes
current inferences and re-runs the rules. Again, not an OWL
feature.
Ps
Another thing I run into is that I get an error on save ("safe failed") after inferencing: the file seems readonly and I have to choose another name...
I assume your data is in the TopBraid project. This project
contains system models that are read-only, hence all files in
that project are read-only. Create a new project (using
New... Project) and add your data files in that project.
-- Scott
-----Original Message-----
From: [email protected] [mailto:[email protected]] On Behalf Of Scott Henninger
Sent: woensdag 21 september 2011 17:52
To: TopBraid Suite Users
Subject: [topbraid-users] Re: spin on rectangle
Michel; In Composer, see see TopBraid/Examples/spinsquare.rdf. That should answer most of your questions.
You don't see constraint violation warnings on multiple values because you haven't defined any that would detect multiple values. The Attribute ss:area : xsd:integer [,1]" in spinsquare.rdf will do this.
Or you can define a constraint, such as:
ASK WHERE
{ ?this :area ?area1 .
?this :area ?area2 .
FILTER (?area1 != ?area2) .
}
I didn't see any instances in your model. When I do add an instance, give it width/length values, and modify the values, I do not see multiple values of area created by incremental inferencing. So to reproduce your scenario, you'll have to provide a step-by-step procedure.
-- Scott
On Sep 21, 10:32 am, "Bohms, H.M. (Michel)" <[email protected]>
wrote:
Hi Scott,
Michel; Let's start by establishing a context. I'll assume you are talking about spinsquare.rdf with SPIN incremental inferencing turned on.
ehhhh, I don't think so, what is spinsquare.rdf? I just have a class
Square in my own ontology, now called rules.ttl as attached.
What version and edition of TBC are you running?
3.5.2.v20110828-2200R
When editing an area or height, I do not see multiple values for area on TBC-ME 3.5.2. So the values are over-ridden by new ones.
when I reun inference I get multiple values (new values are added)
so I thought I add card constraints..but this did nnot help..
The Attribute constraint defined in spin:Rectangle - "Attribute ss:area : xsd:integer [,1]" - will throw a constraint violation warning if there is more than one value for ss:area. Note that constraint violation warnings need to be turned on to run the constraint rules.
I have that "On"
just don't see the constraint flag wheb the are myultiple area values
Thx Michel
-- Scott
On Sep 21, 7:39 am, "Bohms, H.M. (Michel)" <[email protected]>
wrote:
a specific spin question:
when I run inference on diff. width and length I get multiple values
for area also when I define 3 cardinality=1 restrictions (for width,
length and area)
next when I check consistency I see no warnings, is that expected?
isnt there a way to just override de earlier calculated value?
thx Michel
ps
just a general question....as derivations can be seen as specific restriction wouldnt one want only 'restrictions'that are not only checked by also satisfied in some way?
(instead of : construct area:= length*width >>>> ask/make area = length*width, in a symmetric way ...... or is this something for the future??
This e-mail and its contents are subject to the DISCLAIMER
athttp://www.tno.nl/emaildisclaimer
--
You received this message because you are subscribed to the Google Group "TopBraid Suite Users", the topics of which include TopBraid Composer, TopBraid Live, TopBraid Ensemble, SPARQLMotion 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
athttp://groups.google.com/group/topbraid-users?hl=en
rules.ttl
6KViewDownload
rules.ttl.tbc
1KViewDownload
--
You received this message because you are subscribed to the Google Group "TopBraid Suite Users", the topics of which include TopBraid Composer, TopBraid Live, TopBraid Ensemble, SPARQLMotion 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
--
You received this message because you are subscribed to the Google
Group "TopBraid Suite Users", the topics of which include TopBraid
Composer,
TopBraid Live, TopBraid Ensemble, SPARQLMotion 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
--
You received this message because you are subscribed to the Google
Group "TopBraid Suite Users", the topics of which include TopBraid
Composer,
TopBraid Live, TopBraid Ensemble, SPARQLMotion 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
--
You received this message because you are subscribed to the Google
Group "TopBraid Suite Users", the topics of which include TopBraid Composer,
TopBraid Live, TopBraid Ensemble, SPARQLMotion 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
|
