Hi Roberto,
sorry that I need to disagree, but DBpedia is based on Virtuoso, which
has a quad model, i.e. a SQL table with 4 columns.
The sparql query asks for all distinct values in the 3rd column and this
is definitely doable seen from the SQL perspective.
I normally run those queries on a daily basis and I even did the
following query via sparql and it needed around 100 seconds:
Select ?p count(?p) as ?count {?s ?p ?o } group by ?p order by DESC
(?count)
Results are here on 3.5.1:
http://db0.aksw.org/downloads/propertyCount.csv.tar.gz
I did these measurements on a local mirror on one of our servers.
The problem is something else:
- the main store is limited to a result set size of 1000 and has a
limited execution time
- getting a speedy answer depends on the traffic other users produce
- currently there seems to be a bug causing an early transaction time out
The only way to really do statistical measurements is to set up a local
mirror and let queries run against this.
Regards,
Sebastian
On 23.05.2011 18:16, Roberto Mirizzi wrote:
> Il 23/05/2011 17:44, Sarasi Lalithsena ha scritto:
>> Hi,
>>
>> I want to extract all properties from the DBpedia data set and I tried
>> the following query but it did not work. I even tried to limit the
>> results but still it did not work as well.
>>
>> select distinct ?property where {
>> ?s ?property ?o .
>> }
>>
>> I have the properties defined in the DBpedia Ontology but I am
>> wondering whether their can be some other properties not defined in
>> the ontology. If somebody can help me it would be a great help.
> Hi Sarasi,
> probably you will never get the expected results from that query. The
> main problem is that you are asking DBpedia SPARQL endpoint to perform a
> graph matching on all the 672 million triples in DBpedia, and then to
> select distinct property.
> This is too onerous, and the transaction will always time out.
>
>> Thank You
>> Best Regards
>> Sarasi Lalithsena
> best regards,
> roberto
>
--
Dipl. Inf. Sebastian Hellmann
Department of Computer Science, University of Leipzig
Homepage: http://bis.informatik.uni-leipzig.de/SebastianHellmann
Research Group: http://aksw.org
------------------------------------------------------------------------------
What Every C/C++ and Fortran developer Should Know!
Read this article and learn how Intel has extended the reach of its
next-generation tools to help Windows* and Linux* C/C++ and Fortran
developers boost performance applications - including clusters.
http://p.sf.net/sfu/intel-dev2devmay
_______________________________________________
Dbpedia-discussion mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/dbpedia-discussion
