"Could you provide an example. How is a pattern inconsistent? What will be
done to axioms that are potentially conflicting?
I think the first and main issue of this task is to develop a system to
find such axioms. These need to be evaluated and the removal of potential
incorrect axioms will be a postponed step."
For example, A \sub \exists R.B , R's range is C . B and C disjointness.
For such inconsistent pattern type, using the same method to find axioms,
If there's high frequent axioms can be matched all of them.
Or something like, A and B disjoint, C \sub B, C \sub A. C(a). "C(a)" is
to ensure C is not \bot.
Best.
On Tue, Mar 11, 2014 at 5:09 AM, Magnus Knuth <
[email protected]> wrote:
>
> Am 09.03.2014 um 22:06 schrieb Cong Wang:
>
> > Hi Magnus, Heiko and Marco,
>
> Hello Cong Wang,
> and welcome to DBpedia community.
>
> > My name is Cong Wang, a Ph.D. student from Wright State University with
> Pascal Hitzler. I am very interested in the two topics of gsoc 2014,
> "ontolocy consistency check" and "pattern discovery". I found the two
> topics are actually quite interactive, so I write this email to all of you.
> >
> > I had read the related papers. Pattern discovery and type inference are
> both about ontology enrichment (under a maximum likelihood manner) [1,2],
> And while ontology is automatically enriched, there can be inconsistency,
> say, enrichment will cause inconsistency. But on the other hand, enrichment
> will help for detecting potential inconsistency. Since the ontology is too
> big to put into any reasoners, so the idea is remove inconsistency by some
> "suggestions [3]".
> >
> > Now, I have an idea based on your methods. First design a set of
> patterns which itself is inconsistent (the patterns must be high frequent),
> then use some statistical method to extract axioms under these types, if
> there are axioms extracted with high probably, then we may argue there are
> potential conflicts. How do you think of this idea?
>
> Could you provide an example. How is a pattern inconsistent? What will be
> done to axioms that are potentially conflicting?
>
> I think the first and main issue of this task is to develop a system to
> find such axioms. These need to be evaluated and the removal of potential
> incorrect axioms will be a postponed step.
>
> > I also have one question. In [4], I don't understand how the Airpedia
> project relates to consistency checking issues. Can you give me some more
> details?
>
> Afaik Airpedia generates extraction mappings for multiple languages, which
> is a step prior to extraction, so there might be errors in the created
> mappings, but no inconsistencies. We might get inconsistencies in the data
> extracted afterwards. But that should be answered finally by @Marco, since
> I am not involved in this project.
>
> Best regards
> Magnus
>
> > Thanks.
> >
> > [1]. Lorenz Bühmann, Jens Lehmann. Pattern Based Knowledge Base
> Enrichment.
> > [2]. Heiko Paulheim, Christian Bizer: Type Inference on Noisy RDF Data.
> > [3]. Gerald Töpper, Magnus Knuth, Harald Sack. DBpedia Ontology
> Enrichment for Inconsistency Detection
> > [4]. http://wiki.dbpedia.org/gsoc2013/ideas/OntologyCheck?v=l3f
> >
> > Best regards.
> >
> >
> > --
> > Cong Wang
> > Ph.D Candidate,
> > Kno.e.sis Center,
> > Wright State University.
>
>
--
Cong Wang
Ph.D Candidate,
Kno.e.sis Center,
Wright State University.
------------------------------------------------------------------------------
Learn Graph Databases - Download FREE O'Reilly Book
"Graph Databases" is the definitive new guide to graph databases and their
applications. Written by three acclaimed leaders in the field,
this first edition is now available. Download your free book today!
http://p.sf.net/sfu/13534_NeoTech
_______________________________________________
Dbpedia-gsoc mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/dbpedia-gsoc
