I run a query on DBpedia using jena framework, as part of a project, on January 2016 to count the distinct properties of 100 movies. I was writing the report yesterday so I tried to confirm the results, but surprisingly they have changed. On January, I got 100+ properties, yesterday I got 63. I used the same query the same everything!
I even saved the list of properties labels in my hard disk on January that confirm there were 100+. I checked if all of them are still used to describe movies resources on DBpedia, I found that many of them are not. I also had a look at the changelog of their website and this is the only thing I found that may be relevant is on the page (http://wiki.dbpedia.org/dbpedia-version-2016-04) that says:
We now filter out triples from the Raw Infobox Extractor that are already mapped. E.g. no more “ dbo:birthPlace ” and “ dbp:birthPlace|dbp:placeOfBirth|... ” in the same resource. These triples are now moved to the “infobox-properties-mapped” datasets and not loaded on the main endpoint. See issue 22 for more details.
do you think this is the reason, or do you know any explanation to that?
The query I've used :
select (count(distinct ?pr) AS ?count) where {
?su ?pr ?ob .
select ?su{
?su a <http://schema.org/Movie> .
}
limit 100
}
Thanks.
Note that the public service's resource limits (query timeouts, result rowcounts, etc.) apply no matter whether you're using the HTTP-based SPARQL endpoint, Jena/Java/JDBC, ODBC, or otherwise.
Also, I'm quite sure you're not using the query as seen here and now, because it uses invalid syntax.
All that said, you might try your correct query on
(The last certainly hasn't been updated with the new DBpedia data set yet, but it may be at any time, so adopting a disclaimer will be in your best interest. Something like, "This is a publicly-editable, living data set, being extracted from Wikipedia by an evolving codebase, so results may (and will, and have) change over time.")