|
Hi All, I am trying to generate a list of all the universities in Germany, France and Denmark. I built my query below. However, if you run it in dbpedia, you'll notice that it didn't capture all of the universities, and in the results there were even universities in Italy and other countries that we don't want. What may possibly be the issue here? To see the result: Here
|
|
Hi, there's nothing wrong with your query. However, based on the DBpedia data, there can be universities that are located in many countries (international universities). For instance, this university is located in Germany, Italy and France. thanks for the quick response, however, many universities didn't even make it to the result. like university of paderborn in germany http://dbpedia.org/page/University_of_Paderborn here is another uni in france that didn't make it to the results: http://dbpedia.org/page/Montpellier_2_University Because you only specified the types to be either http://dbpedia.org/class/yago/University108286163 or http://dbpedia.org/resource/Public_university. If you look at the data of Paderborn, especially in the values of rdf:type property, you'd understand why. do you mind to elaborate? we switched to http://schema.org/CollegeOrUniversity, we get much more results, but still not all of the results (e.g. Paderborn). I have also updated my query above |
|
To add to @fadirra's answer, and to ask the more general underlying question, DBpedia is neither 100% correct nor 100% complete nor 100% "consistent" (in terms of use of vocabulary):
In other words, don't assume that queries against DBpedia will always return all and only correct results, and don't assume that a specific structure of query will hit upon all possible heterogeneities in how the target data are described. If you feel this isn't good enough, I guess keeping an eye on Wikidata developments might be a good idea. ;) I should add, this is not a criticism of the great ongoing work over at DBpedia. This is just a reflection of how difficult a task it is to automatically extract comprehensive, high quality RDF data from a source like Wikipedia. |

