I find this rdf:type in the latest dumps, <http://dbpedia.org/resource/Hidehiko_Shimizu> <http://www.w3.org/1999/02/22-rdf-syntax-ns#type> <http://dbpedia.org/ontology/SoccerManager>
but he doesn't have a label [p...@haruhi dbpedia_3.5.1]$ bzgrep Hidehiko_Shimizu labels_en.nt.bz2 [p...@haruhi dbpedia_3.5.1]$ There's a page about this guy in Wikipedia that looks pretty normal, except for the fact that the deletionists want it to go away: http://en.wikipedia.org/wiki/Hidehiko_Shimizu What's up here? It seems to me that there ought to be some kind of acceptance tests done on the dumps so that we know the key structure makes sense. For the last few versions of dbpedia, it's been (mostly) true that assertions are only made about things that (i) have a label, or (ii) are the subject of a redirect. The exception to that has been that there are wikilinks to pages that don't exist, precisely because there ~are~ wikilinks to pages that don't exist. I can see that some good may come out of extracting Engines out of automobile descriptions and PersonFunctions out of persons, but this one just looks like a glitch. ------------------------------------------------------------------------------ _______________________________________________ Dbpedia-discussion mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/dbpedia-discussion
