TopBraid Composer Users Google Group

Re: [tbc-users] Re: Auto add spin subclass inferences

hol...@topquadrant.com (Holger Knublauch) — Mon, 21 Dec 2009 12:21:30 UT

Sure. Of course any DB-backed solution will be slower than an in-memory one. You can work around some slowness with careful engineering. For example,
- compute the class-to-rule mapping only once (walking the SPIN RDF syntax expressions might be slow on a DB).
- if there are no cycles in your rules, then the inferred triples can be written to an in-memory model to prevent many small update requests.

Re: [tbc-users] Re: Auto add spin subclass inferences

j...@netbrackets.com (Jeffrey Schmitz) — Sun, 20 Dec 2009 23:49:26 UT

Just a quick point before I look into how to use union graphs.
The SDB backed model I was using wasn't an OntModel, it was just
produced by copying the OntModel into it. e.g.
_store = SDBFactory.connectStore("sdb.t tl");
Model _sdbModel =
SDBFactory.connectNamedModel(_ store, URI);

Re: [tbc-users] registerAllSPINModules exception

j...@netbrackets.com (Jeffrey Schmitz) — Sun, 20 Dec 2009 13:59:09 UT

The problem seems to be that I wasn't following the note:
public void registerAll(Model model)
Registers all functions and templates from a given Model. Note that
the Model should be an OntModel that also imports the system
namespaces spin.owl and sp.owl - otherwise the system may not be able
to transform the SPIN RDF into the correct SPARQL string.

Re: [tbc-users] registerAllSPINModules exception

hol...@topquadrant.com (Holger Knublauch) — Sun, 20 Dec 2009 07:32:53 UT

Hi Jeff,
could you send me the spl.spin.owl file from your distribution? Also, which Jena version are you using?
Thanks
Holger

Re: SparqlMotion script not retrieving all subclasses from an ontology that imports other ontologies

shennin...@topquadrant.com (Scott Henninger) — Sat, 19 Dec 2009 18:28:08 UT

TK; It's not entirely clear from the description what the count is
for, but using an aggregate may be an effective way to do this. The
following gets all subclasses of owl:Thing and a count of the number
of subclasses.
SELECT ?cls (count(?subc) as ?subClasses)
WHERE {
?cls rdfs:subClassOf+ owl:Thing .

registerAllSPINModules exception

j...@netbrackets.com (Jeffrey Schmitz) — Sat, 19 Dec 2009 15:06:29 UT

Hello,
I'm getting an exception on the call to registerAllSPINModules
(stack trace below). Stepping through the code, it's happeining on
the following spin:Function defined in the spl ontology (spl.spin.owl):
[link]
I'm not quite sure how to debug this one. I believe I have the latest

Re: [tbc-users] Re: Auto add spin subclass inferences

hol...@topquadrant.com (Holger Knublauch) — Sat, 19 Dec 2009 13:08:19 UT

Jeff,
whenever you have a MultiUnion graph (or OntModel) that consists of more than one sub-graph, then the performance of the SPARQL engine might go down significantly. This is because the triple matches may need to dynamically merge partial results from multiple sub-graphs. On the other hand, if you just have a single graph (incl a single SDB graph), then the system can exploit native optimizations and do complex graph patterns with a single, optimized operation. In practice my guess is that you will have best performance if you put all sub-graphs into the same SDB (possibly split into named graphs) and then operate on the union graph (via the named graph <urn:x-arq:UnionGraph>).

Re: [tbc-users] Re: Auto add spin subclass inferences

j...@netbrackets.com (Jeffrey Schmitz) — Sat, 19 Dec 2009 05:40:02 UT

Well, I was doing things so fast and furiously that I'm not for sure
what it was that helped, but I wasn't tweaking any of my queries so it
wasn't anything like that. I think it had mostly to do with making
sure I was using an "in memory" model instead of a database backed
model. When I created an Ontology model like:

Re: [tbc-users] SparqlMotion script not retrieving all subclasses from an ontology that imports other ontologies

hol...@topquadrant.com (Holger Knublauch) — Sat, 19 Dec 2009 05:23:35 UT

Please make sure that the file explicitly asserts all rdfs:subClassOf triples. When you open files in TBC, the system will add "trivial" inferences to make sure that each named class has at least one named superclass. These are then inferred triples, which would not be available by default if your file is loaded through SPARQLMotion. Also, you may want to import the system triples to get the same environment like in TBC.

Re: [tbc-users] Re: Auto add spin subclass inferences

hol...@topquadrant.com (Holger Knublauch) — Sat, 19 Dec 2009 05:14:48 UT

I am glad to hear that. If you have any lessons learnt on the "dumb things" then we would appreciate if you could share them with others on this mailing list. For example, changing the order of clauses can reduce many orders of magnitude off query execution times, e.g. see
[link]

Re: [tbc-users] Magic Properties vs Spin Rules

hol...@topquadrant.com (Holger Knublauch) — Sat, 19 Dec 2009 05:07:28 UT

Hi Arthur,
by default, magic properties are not available as "inferred" triples while browsing a model in TopBraid. They are computed on demand in SPARQL queries only. However, you can add a SPIN rule such as
CONSTRUCT {
?a my:magicProperty ?b .
WHERE {
?a my:magicProperty ?b .
to make them explicit. This way you avoid duplicating the definitions of the body. I am sure this could be generalized into a single rule for all magic properties, at least with a bit of Java coding (e.g. write a meta magic property that runs all other magic properties). If that's of interest I can add something to 3.3

SparqlMotion script not retrieving all subclasses from an ontology that imports other ontologies

thomas.carpen...@ctg.com (TKC) — Fri, 18 Dec 2009 22:31:55 UT

I have written a SparqlMotion script that imports an ontology, say
"Main," using the "Import RDF From Workspace" module. (Now, "Main,"
itself, imports several sub-ontologies.) Following this module, I have
a Sparql query that attempts to retrieve all subclasses of
"owl:Thing"--from "Main" and from all ontologies that "Main" imports.

Magic Properties vs Spin Rules

arthur.a.k...@gmail.com (Arthur Keen) — Sat, 19 Dec 2009 00:37:23 UT

I am looking into SPIN to compute derived quantities in an analytical model. I studied the blog on magic properties, and tried it out the grandParent magic property on kennedys.owl and it works as promised. I understand that these magic properties are intended to extend SPARQL, however, since they are properties, could they be extended to be used for late binding properties on objects, so that the values could be computed on access? I tried unsuccessfully to do this, and concluded that were not meant to be used in this way and that I should use spin rules for computing derived properties on objects. Is that correct?

Re: [tbc-users] Re: Auto add spin subclass inferences

j...@netbrackets.com (Jeffrey Schmitz) — Fri, 18 Dec 2009 20:47:56 UT

Actually, after fixing a few dumb things I was doing, the speed has
improved immensely. In the end, I don't think supressing the spin
model inferences really buys much since inferences are now running in
under a second, even with those in.
Jeff

Re: [tbc-users] Re: Auto add spin subclass inferences

j...@netbrackets.com (Jeffrey Schmitz) — Fri, 18 Dec 2009 20:05:03 UT

Thanks for the suggestions, just a couple comments below...
There are some class/subclass hierarchies, but currently very little
instance data (that will change). Looking at the inferences produced,
most of the them are from the spin models themselves. e.g.
[[link], [link]