query-4f7a4e7949913c0217f2322a9c3a0417

rq turtle/ttl

most Gender-imbalanced occupations in odia wikipedia SELECT ?occupation ?occupationLabel ?women ?men WITH { SELECT ?occupation (SUM(?woman) AS ?women) (SUM(?man) AS ?men) WHERE { #BIND(wd:Q668 AS ?country). # substitute country of your choice here (remove the Q668, hit Ctrl+Space and start typing the country name, select the right one from the suggestions with the arrow keys and insert it with Enter) ?person wdt:P31 wd:Q5;. #?person wdt:P27 ?country. # remove this to search world-wide – results in query timeout unless you reduce the dataset in some other fashion ?person p:P106/ps:P106 ?occupation. ?article schema:about ?person . ?article schema:isPartOf https://or.wikipedia.org/ . { ?person wdt:P21 wd:Q6581097. BIND(1 AS ?man). } UNION { ?person wdt:P21 wd:Q2449503. BIND(1 AS ?man). } UNION { ?person wdt:P21 wd:Q6581072. BIND(1 AS ?woman). } UNION { ?person wdt:P21 wd:Q1052281. BIND(1 AS ?woman). } # TODO consider how to count P21: Q48270 (genderqueer) and Q1097630 (intersex) } GROUP BY ?occupation } AS %results WHERE { INCLUDE %results. # subquery as optimization: don’t make the label service fetch labels for every ?person in the subquery SERVICE wikibase:label { bd:serviceParam wikibase:language "[AUTO_LANGUAGE],en". } } ORDER BY

divide ?variable+1 instead of ?variable to avoid division by zero

# (unfortunately, we can’t use ABS(LOG(x/y)) instead of the IF() because SPARQL has no LOG function) DESC(IF(?women > ?men, (?women+1)/(?men+1), (?men+1)/(?women+1))) # if the ratio is the same, show results with more people in total first DESC(?women + ?men)

Use at

• https://query.wikidata.org/sparql

PREFIX wikibase: <http://wikiba.se/ontology#>
PREFIX wdt: <http://www.wikidata.org/prop/direct/>
PREFIX wd: <http://www.wikidata.org/entity/>
PREFIX schema: <http://schema.org/>
PREFIX ps: <http://www.wikidata.org/prop/statement/>
PREFIX p: <http://www.wikidata.org/prop/>
PREFIX bd: <http://www.bigdata.com/rdf#>
# most Gender-imbalanced occupations in odia wikipedia 
SELECT ?occupation ?occupationLabel ?women ?men WHERE {
   {
  SELECT ?occupation (SUM(?woman) AS ?women) (SUM(?man) AS ?men) WHERE {
    #BIND(wd:Q668 AS ?country). # substitute country of your choice here (remove the Q668, hit Ctrl+Space and start typing the country name, select the right one from the suggestions with the arrow keys and insert it with Enter)
    ?person wdt:P31 wd:Q5;.
    #?person wdt:P27 ?country. # remove this to search world-wide – results in query timeout unless you reduce the dataset in some other fashion
     ?person p:P106/ps:P106 ?occupation.
?article schema:about ?person . ?article schema:isPartOf <https://or.wikipedia.org/> .
    { ?person wdt:P21 wd:Q6581097. BIND(1 AS ?man). } UNION
    { ?person wdt:P21 wd:Q2449503. BIND(1 AS ?man). } UNION
    { ?person wdt:P21 wd:Q6581072. BIND(1 AS ?woman). } UNION
    { ?person wdt:P21 wd:Q1052281. BIND(1 AS ?woman). }
    # TODO consider how to count P21: Q48270 (genderqueer) and Q1097630 (intersex)
  }
  GROUP BY ?occupation
} # subquery as optimization: don’t make the label service fetch labels for every ?person in the subquery
  SERVICE wikibase:label { bd:serviceParam wikibase:language "[AUTO_LANGUAGE],en". }
}
ORDER BY
# divide ?variable+1 instead of ?variable to avoid division by zero
  # (unfortunately, we can’t use ABS(LOG(x/y)) instead of the IF() because SPARQL has no LOG function)
  DESC(IF(?women > ?men, (?women+1)/(?men+1), (?men+1)/(?women+1)))
  # if the ratio is the same, show results with more people in total first
  DESC(?women + ?men)

Query found at

• https://www.wikidata.org/wiki/User:MKar/Query