You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@jena.apache.org by Markus Freudenberg <ma...@gmail.com> on 2016/02/24 18:41:09 UTC

Re: [Dbpedia-discussion] Dbo vs Dbp

Hello Kumar,

we differentiate between information extracted from the wikipedia dumps
without an alignment to the DBpedia ontology (raw extraction) and the
mapping based extraction (based on mappings between wikipedia infoboxes and
the ontology).


   -

   http://dbpedia.org/property/ (prefix dbp) for representing properties
   extracted from the raw infobox extraction
   -

   http://dbpedia.org/ontology/ (prefix dbo) for representing the DBpedia
   ontology

I also refer you to the description of the mapping-based properties dataset
(containing the mapping based properties...) :
Mapping-based Properties

*High-quality data extracted from Infoboxes using the mapping-based
extraction. The predicates in this dataset are in the /ontology/ namespace.*
Note that this data is of much higher quality than the Raw Infobox
Properties in the /property/ namespace. For example, there are three
different raw Wikipedia infobox properties for the birth date of a person.
In the the /ontology/ namespace, they are all *mapped onto one relation*
http://dbpedia.org/ontology/birthDate. It is a strong point of DBpedia
to unify these relations.

I hope this is helpful,

Markus

On Wed, Feb 24, 2016 at 5:46 PM, kumar rohit <ku...@gmail.com> wrote:

> Hello
>
> I do not know when to use dbo and when to use dbp when we query using
> dbpedia. Kindly if some one can explain how to use these in which scenarios?
>
> Regards
>
>
> ------------------------------------------------------------------------------
> Site24x7 APM Insight: Get Deep Visibility into Application Performance
> APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month
> Monitor end-to-end web transactions and take corrective actions now
> Troubleshoot faster and improve end-user experience. Signup Now!
> http://pubads.g.doubleclick.net/gampad/clk?id=272487151&iu=/4140
> _______________________________________________
> Dbpedia-discussion mailing list
> Dbpedia-discussion@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/dbpedia-discussion
>
>

RE: [Dbpedia-discussion] Dbo vs Dbp

Posted by Vladimir Alexiev <vl...@ontotext.com>.
Kumar:
- if a property you need is in dbo: use that. But double that for the objects in interest, it is indeed present in most of them
- if a property you need is not in dbo: or many of the objects of interest are missing it, use dbp: .
  As described by Markus, you may need to use several dbp: props, especially across different language DBpedias: each dbp: is named in the local language

> For example, there are three different raw Wikipedia infobox properties for the birth date of a person.
> In the the /ontology/ namespace, they are all mapped onto one relation 

In addition to the merging of several dbp: into one dbo:, there is another advantage of dbo:
- dbp: values of an array of subobjects are not correlated between each other.
  But if the mapping is properly modeled (e.g. using IntermediateNodeMapping), they can be properly correlated in dbo:

> data is of much higher quality than the Raw Infobox Properties in the /property/ namespace.

That is the theory anyway :-)

But dbp: come about automatically: each template property is emitted as a corresponding dbo: property
In contrast, dbo: only come about, when someone has made an appropriate mapping.

There are at least 2 potential problems with dbo: that a consumer needs to be aware of:
- dbp:birthDate may be mapped to dbo:birthDate for 100 templates, but missing for the 101st template (or the mapping for that template may be missing altogether).
  How would a consumer detect this problem?
- each dbo: is either an ObjectProperty or DatatypeProperty.
  This means that each returns only URLs, or only literals, but not both.
  dbp: doesn't have such restriction.
  Search for "dichotomy" in http://vladimiralexiev.github.io/pres/20150209-dbpedia/dbpedia-problems-long.html

Cheers! Vladimir