query-1ee14a72ac44396a957ec1c853d24c3e

rq turtle/ttl

When we run this query now (2020-11) we get ca. 5K hits. In 2020-09 we ran "the same query" and got 13K hits. We have the Wikidata SPARQL-XML output from 2020-09 which indeed has 13K hits. It's a superset of our current 5K hits. But it doesn't record what query we submitted! Some of us are sure we submitted the same query. Others think that the query was a broader one. So one way to resolve it would have been to re-run it with 2020-09 version. 09:53, 26 November 2020 (UTC)) talk (PetermrThe wider issue is whether it's possible to record the query in the SPARQL-XML. I think that would be a useful feature, but it's probably non-trivial.09:57, 26 November 2020 (UTC)) talk (SteakYou could just reverse engineer from the results list, what the query must have been. 09:59, 26 November 2020 (UTC) JuraSome bot removed plenty of P31 statements the other day. It might be worth investigating if that was correct. --- 10:45, 26 November 2020 (UTC)) talk (Dipsacus fullonum with best rank, but that they don't have the statement any longer or it is not best rank any longer. --(Q12136)disease (P31)instance of : You could try to find some items which were in your old results but not in the new results. Then look at the history for these items to confirm that they once had the statement Petermr@ 10:14, 26 November 2020 (UTC) JuraThe problem is that I didn't find the same identifiers added elsewhere. The nice thing about the field is that it doesn't really overlap with things that interest me. --- 13:43, 26 November 2020 (UTC) Jura. --- Property_talk:P699Sounds sensible. Maybe worth summarizing on 13:27, 26 November 2020 (UTC)) talk (Andrawaag, we minted both a P31 and P279 at the time. However, semantically that was incorrect and not in line with how it is expressed in the source. There everything is connecting as subclasses. This task is long overdue, since the WDQS already exists for a while, but since we are updating identifier mappings we are also removing those redundant (and partly incorrect) P31 claims. So the number returning on the query above might even decrease more in the near future. The suggestion to use wdt:P31/wdt:P279 should return all diseases. --(Q5282129)Disease Ontology . Regarding the P31. Before WDQS existed there was WDQ which made traversing a tree rather complex. To be able to query for all disease items sourced from the (Q5282129)Disease Ontology to express exact similarity. Since XREFs mappings were no longer accurate we had to remove them from Wikidata and replace them with those that were expressed using skos:exactMatch. So indeed identifiers were removed (but only those with a reference to the skos:exactMatch started using (Q5282129)Disease Ontology only used Xrefs to express mappings to external knowledge bases and identifier. However this turned out to be problematic, since the semantics of Xrefs are not wel defined. Not every XREF mapping is an exact match. Recently, (Q5282129)Disease Ontology . Until recently, (Q27468140)Mondo Disease Ontology and (Q5282129)Disease Ontology : There are two things hapening here. The bot is aligning Wikidata with Tagishsimon@ 10:12, 26 November 2020 (UTC)) talk (Tagishsimon: what's the story here? --Sulhasan, AndrawaagOne presumes its owners know their business. @ 10:09, 26 November 2020 (UTC) Jura --- https://www.wikidata.org/w/index.php?title=Q1134891&action=historyThe bot also removes several identifiers: 10:03, 26 November 2020 (UTC)) talk (Tagishsimon ... if so the solution is probably to query the disease subclass tree - wdt:P31/wdt:P279 --https://www.wikidata.org/w/index.php?title=Q17540023&action=historyExample of P31 removal here Thanks, that's close to the answer, I think. Adding P279* gives ca 16K hits (16745 results in 10411 ms)

Use at

PREFIX wikibase: <http://wikiba.se/ontology#>
PREFIX wdt: <http://www.wikidata.org/prop/direct/>
PREFIX wd: <http://www.wikidata.org/entity/>
PREFIX bd: <http://www.bigdata.com/rdf#>
SELECT ?item ?itemLabel 
WHERE 
{
  ?item wdt:P31/wdt:P279* wd:Q12136 .
  SERVICE wikibase:label { bd:serviceParam wikibase:language "[AUTO_LANGUAGE],en". } 
}

Query found at

graph TD classDef projected fill:lightgreen; classDef literal fill:orange; classDef iri fill:yellow; v1("?item"):::projected a1((" ")) c5(["bd:serviceParam"]):::iri c7(["#91;AUTO_LANGUAGE#93;,en"]):::literal c3(["wd:Q12136"]):::iri v1 --"wdt:P31"--> a1 a1 --"wdt:P279"--> c3 subgraph s1["http://wikiba.se/ontology#label"] style s1 stroke-width:4px; c5 --"wikibase:language"--> c7 end