[Kim-discussion] runtime error as KIM Server is starting
Marin Nozhchev
mnozchev at sirma.bg
Fri Jul 16 08:20:27 EDT 2010
Hi Chris,
You only need to add a stage to your GATE pipeline that imports new
named entities with their canonical name into the KIM Semantic
Repository . That can be achieved using the Java RMI API of the KIM
Server. Does that work for your usecase ?
I'll send the relevant links after we release the full documentation.
Best,
Marin
On 14.07.2010 18:09 ч., Chris Baker wrote:
> Hi Marin,
> If I understand this correctly our knowledgebase must be instantiated
> with
> canonical names of all instances that could possibly exist for our
> domain model.
> In our scenario we instantiate canonical names only when they are
> found in the
> corpus by text mining. So this is not intuitive to us.
> Can you elaborate.
> Chris
>
> On Wed, Jul 14, 2010 at 11:54 AM, Chris Baker <denguehost at gmail.com
> <mailto:denguehost at gmail.com>> wrote:
>
> Hi Marin,
> Does this mean that we have to create Alias classes into our
> own ontology and then map our alias class with ProtonS Alias class ?
> Should our Alias class be a subclass of Proton Alias class.
> Should we use SameAs or Subclass
> I am also concerened that all out synonyms in our gazateer
> now need to be represented in our ontology. Bascially an RDF
> gazeteer. Is this correct ?
> Chris and Alex
> On Wed, Jul 14, 2010 at 11:48 AM, Marin Nozhchev
> <marin.nozhchev at ontotext.com <mailto:marin.nozhchev at ontotext.com>>
> wrote:
>
> Hi Alex,
>
> The alias definitions and the individual declaration must be
> in the same place. In this case, everything should go in your
> test ontology. There should be no reason to edit any of the
> PROTON files.
>
> I'm sorry if the documentation confused you to think that the
> links between entities and aliases and the aliases should be
> added to PROTON System. I CC'ed one of our editor, who will
> review the relevant text*
>
> Hope this helps,
> Marin
>
> *
> http://nmwiki.ontotext.com/nmwiki/pmwiki.php/KimDocs-30-EN/ExtendInformationExtraction
>
>
>
> On 14.07.2010 17:23 ч., alexk at unb.ca <mailto:alexk at unb.ca> wrote:
>
> Hi Marin,
>
> Let' say I have individual indA_1 of the class A. This
> individual is included in my test ontology.
>
> Now I am trying to follow your documentation and have
> create individual indA_1a of the class Allias. This
> individual is added to the Protons ontology. My
> unserstanding is that I have connect
> indA_1:hasAlias:indA_1a
>
> The problem is hasAlias:indA_1a belongs to Protons while
> indA_1 belongs to our test ontology.
>
> The ontologies are not connect by import (only by mapping
> file).
>
> So, where I have to place the statement like
> indA_1:hasAlias:indA_1a ?
>
>
> Thank you
>
> Alex
>
>
>
>
>
> Quoting Marin Nozhchev <marin.nozhchev at ontotext.com
> <mailto:marin.nozhchev at ontotext.com>>:
>
> Hi Alex,
>
> In the Ontology screen, the line: "A, a List, a
> Restriction" means that A is class of type (owl:)List
> and (owl:)Restriction. Thus clicking on the
> Restriction, thus shows the known metadata about
> owl:Restriction.
>
> The KIM UI doesn't show what restrictions apply to
> class A.
>
> Did I understand you correctly?
>
> Best,
> Marin
>
> On 13.07.2010 17:47 ч., alexk at unb.ca
> <mailto:alexk at unb.ca> wrote:
>
> Yes, it seems to work now. Thank you.
>
> Only concern I have at the moment is that I click
> on restrictions for class A or B (Ontology screen)
> I got a list of data properties that are not in my
> Ontology. May be they are inherited? Is it what
> expected to be?
>
>
> My next step will be to populate it with instances
> and try searching.
>
>
> Quoting Marin Nozhchev
> <marin.nozhchev at ontotext.com
> <mailto:marin.nozhchev at ontotext.com>>:
>
> I tested that visibility file and it is
> correct. Please tell me if any other issues
> remain after applying the patch.
>
> Best,
> Marin
>
> On 13.07.2010 17:20 ч., alexk at unb.ca
> <mailto:alexk at unb.ca> wrote:
>
> visibility file attached...
>
>
> Quoting Marin Nozhchev
> <marin.nozhchev at ontotext.com
> <mailto:marin.nozhchev at ontotext.com>>:
>
> Hi Alex, all,
>
> The attached patch that updates KIM
> Platform 3.0-RC3 to 3.0-RC4. Unzip it
> in the KIM installation folder, while
> the KIM Server and Apache Tomcat (or
> other servlet container that you are
> using) are stopped.
>
> I have tested that your test ontology
> loads and displays correctly after
> patching. You don't need to reset KIM
> by deleting populated after patching.
>
> Best,
> Marin
>
> P.S. Matt, 3.0-RC4 will be available
> on the ontotext.com/kim
> <http://ontotext.com/kim> later today.
> On 12.07.2010 18:32 ч., Marin Nozhchev
> wrote:
>
> Hi Alex, All,
>
> Thank you for reporting the issue.
> It looks like this is an issue in
> KIM, which fails to account for
> RDF blank nodes that appear as a
> result of OWL inference over some
> ontologies.
>
> This issue was aggravated by a
> recent change in BigOWLIM. We will
> fix that immediately, as other
> users may experience that too.
> I'll send you a patch tomorrow and
> the team will put a new release
> candidate with the patch for download.
>
> Best,
> Marin
>
> On 12.07.2010 18:06 ч.,
> alexk at unb.ca <mailto:alexk at unb.ca>
> wrote:
>
>
> Hi Marin,
>
> We tried to integrate our
> ontology with Kim. (It is
> about small test ontology,
> just to learn how to do it).
> We got a Java runtime error as
> KIM Server is starting (please
> see the error message in the
> attachment). The ontology,
> mapping and configuration are
> also attached. Would you have
> a minute to take a look on it?
>
> Any suggestions?
>
> Thank you
>
> Alex
>
>
> Alexandre Kouznetsov, PHD
> postdoctoral research scientist
> Knowledge Navigation
> Infrastructure Team
> Dept. Computer Science &
> Applied Statistics
> University of New Brunswick,
> Canada
>
> tel: 506 6399358
> emails: akouz086 at uottawa.ca
> <mailto:akouz086 at uottawa.ca> ,
> alexk at unb.ca <mailto:alexk at unb.ca>
>
> http://www.unbsj.ca/sase/csas/facultyPage.php?member=Dr.%20Alex%20Kouznetsov
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
> --
> Christopher J. O. Baker Ph. D.
> Associate Professor
> Dept. Computer Science and Applied Statistics
> University of New Brunswick, Canada
> http://ca.linkedin.com/in/christopherjobaker
>
>
>
>
> --
> Christopher J. O. Baker Ph. D.
> Associate Professor
> Dept. Computer Science and Applied Statistics
> University of New Brunswick, Canada
> http://ca.linkedin.com/in/christopherjobaker
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://ontotext.com/pipermail/kim-discussion/attachments/20100716/26d40e56/attachment-0001.html>
More information about the Kim-discussion
mailing list