> How is 40000 records out of almost 100000 a good result? Dbpedia resources have multiple rdfs:label, one for every language. You need to use a filter to only get @en triples.
Georgi -- Georgi Kobilarov Freie Universität Berlin www.georgikobilarov.com > -----Original Message----- > From: Michael Haas [mailto:[email protected]] > Sent: Friday, July 03, 2009 11:54 PM > To: Kingsley Idehen > Cc: Georgi Kobilarov; DBpedia > Subject: Re: [Dbpedia-discussion] Current state of the SPARQL endpoint > > Kingsley Idehen wrote: > > SELECT count(distinct ?name) > > WHERE { ?a a <http://dbpedia.org/ontology/Actor> . > > ?a <http://www.w3.org/2000/01/rdf-schema#label> > ?name > > } > > > > Gives you a total of: 98,641 > > > <snip> > > > > > We have a window of 40,000 records > > > > > > How is 40000 records out of almost 100000 a good result? Especially > when > obtaining large result sets was no problem two weeks ago? > > > (amply generous since you are clearly > > crawling this data). > > Is that not a valid use case? Queries similar to "Give me all actors" > are even mentioned on dbpedia.org [0]. In any case, extracting data for > (computational) linguistic purposes sounds like a good use case for an > ontology built from Wikipedia. > > Also, accusing me of "crawling" this data is a bit ironic considering > how future DBpedia versions will be built. > > > Regards, > > Michael > > > [0] http://wiki.dbpedia.org/OnlineAccess#h28-4 ------------------------------------------------------------------------------ _______________________________________________ Dbpedia-discussion mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/dbpedia-discussion
