Wikidata:ScienceSource project/Queries

This page is for sample ScienceSource SPARQL queries. Appearances to the contrary, these are not queries for the Wikidata Query Service. They are to be run on

http://sciencesource-query.wmflabs.org/

Links are provided on a ScienceSource wiki page.

The data schema at

http://sciencesource.wmflabs.org/wiki/Data_schema

is in use, so the P- and Q-numbers do not refer to those in use on Wikidata. The queries are displayed here using {{Sparql}} for readability.

For resources on SPARQL and homegrown queries, there are numerous examples on User:Charles Matthews/Queries and in Wikidata:WikiFactMine/Core SPARQL.

Articles by ascending length

• #Articles by ascending notional length
  SELECT ?annotation ?title ?length ?article
  WHERE 
     {?annotation wdt:P19 ?anchor.
      ?anchor wdt:P7 wd:Q6.
      ?anchor wdt:P12 ?article.
      ?anchor wdt:P10 ?length.
      ?article wdt:P20 ?title
     }
  ORDER BY ASC (?length)
  

  Try it!

Annotations for a particular term

• #Find all annotations with term "toxocariasis", showing the article in which the term is found.
  SELECT ?item ?article ?term 
    WHERE {?item wdt:P3 wd:Q5.
           ?item wdt:P15 ?term.
           ?item wdt:P20 ?article.
          FILTER (?term = "toxocariasis")
           }
  

  Try it!

Distinct terms found

• #Show all distinct terms found (currently disease and drug terms)
  SELECT DISTINCT ?term
    WHERE {?item wdt:P3 wd:Q5.
           ?item wdt:P15 ?term}
  

  Try it!

Counting annotations

• #Counting drug annotations
  #defaultView:BubbleChart
  SELECT ?drug ?drugLabel ?count
  WHERE 
  {
    {
      SELECT ?drug (COUNT(?annotation) AS ?count)    
  WHERE {
          ?annotation wdt:P15 ?drug.
          ?annotation wdt:P16 "infectiousdiseasesdrugs" .
         }
    GROUP BY ?drug 
    HAVING (?count > 1)        
    }
     SERVICE wikibase:label { bd:serviceParam wikibase:language "en" }
  }
  ORDER BY DESC(?count)
  LIMIT 500
  

  Try it!

• #Counting disease annotations
  #defaultView:BubbleChart
  SELECT ?disease ?diseaseLabel ?count
  WHERE 
  {
    {
      SELECT ?disease (COUNT(?annotation) AS ?count)    
  WHERE {
          ?annotation wdt:P15 ?disease.
          ?annotation wdt:P16 "infectiousdiseases" .
         }
    GROUP BY ?disease
    HAVING (?count > 1)        
    }
     SERVICE wikibase:label { bd:serviceParam wikibase:language "en" }
  }
  ORDER BY DESC(?count)
  LIMIT 500
  

  Try it!

Co-occurrence 1

• #Co-occurrence of drug and disease terms (duplications removed)
  SELECT DISTINCT ?articletitle ?disease ?drug
    WHERE {?term1 wdt:P19 ?anchor1.
           ?term2 wdt:P19 ?anchor2.
           ?term1 wdt:P15 ?disease.
           ?term2 wdt:P15 ?drug.
           ?term1 wdt:P16 "infectiousdiseases".
           ?term2 wdt:P16 "infectiousdiseasesdrugs".
           ?anchor1 wdt:P12 ?article.
           ?anchor2 wdt:P12 ?article.
           ?article wdt:P20 ?articletitle.
           SERVICE wikibase:label { bd:serviceParam wikibase:language "[AUTO_LANGUAGE],en". }
           }
  

  Try it!

1. #Co-occurrence filtered by the condition "disease term before drug term"
   SELECT ?articletitle ?disease ?drug ?charnumber1 ?charnumber2
     WHERE {?term1 wdt:P19 ?anchor1.
            ?term2 wdt:P19 ?anchor2.
            ?term1 wdt:P15 ?disease.
            ?term2 wdt:P15 ?drug.
            ?term1 wdt:P16 "infectiousdiseases".
            ?term2 wdt:P16 "infectiousdiseasesdrugs".
            ?anchor1 wdt:P10 ?charnumber1.
            ?anchor2 wdt:P10 ?charnumber2.
            ?anchor1 wdt:P12 ?article.
            ?anchor2 wdt:P12 ?article.
            ?article wdt:P20 ?articletitle.
            FILTER (?charnumber2 > ?charnumber1)
            SERVICE wikibase:label { bd:serviceParam wikibase:language "[AUTO_LANGUAGE],en". }
            }
   

   Try it!

Co-occurrence 2

• #Co-occurrence filtered by the condition "disease term before drug term" and character distance apart < 100
  SELECT ?articletitle ?disease ?drug ?charnumber1 ?charnumber2
    WHERE {?term1 wdt:P19 ?anchor1.
           ?term2 wdt:P19 ?anchor2.
           ?term1 wdt:P15 ?disease.
           ?term2 wdt:P15 ?drug.
           ?term1 wdt:P16 "infectiousdiseases".
           ?term2 wdt:P16 "infectiousdiseasesdrugs".
           ?anchor1 wdt:P10 ?charnumber1.
           ?anchor2 wdt:P10 ?charnumber2.
           ?anchor1 wdt:P12 ?article.
           ?anchor2 wdt:P12 ?article.
           ?article wdt:P20 ?articletitle.
           FILTER (?charnumber2 > ?charnumber1)
           FILTER (?charnumber2 - ?charnumber1) < 100)
           SERVICE wikibase:label { bd:serviceParam wikibase:language "[AUTO_LANGUAGE],en". }
           }
  

  Try it!

Co-occurrence 3

• #Co-occurrence (filtered) displayed for single article Q6679
  #defaultView:Dimensions
  SELECT  ?drugLabel ?charnumber2 ?charnumber1 ?diseaseLabel
  WHERE {
           ?anchor1 wdt:P12 wd:Q6679;
                    wdt:P10 ?charnumber1.
           ?anchor2 wdt:P12 wd:Q6679;
                    wdt:P10 ?charnumber2.
           ?term1 wdt:P19 ?anchor1.
           ?term2 wdt:P19 ?anchor2.
           ?term1 wdt:P15 ?disease.
           ?term2 wdt:P15 ?drug.
           ?term1 wdt:P16 "infectiousdiseases".
           ?term2 wdt:P16 "infectiousdiseasesdrugs".     
           FILTER (?charnumber2 > ?charnumber1)
           FILTER (?charnumber2 - ?charnumber1 < 200)
   
    SERVICE wikibase:label { bd:serviceParam wikibase:language "en". }
  }
  

  Try it!

Counting annotations in a time period

1. #Counting annotations in a time period: publication date 2013 earliest
   SELECT ?annotation ?term ?title ?date1
   WHERE 
      {?annotation wdt:P19 ?anchor.
       ?annotation wdt:P15 ?term.
       ?anchor wdt:P12 ?article.
       ?article wdt:P17 ?date.
       ?article wdt:P20 ?title.
      BIND (YEAR(?date) AS ?date1)
      FILTER (?date1 > 2012)
      }
   

   Try it!

Annotations and quantiles

• #Annotations in notional first 10% of paper, ordered by absolute distance
  #Good correlation with terms in title
  SELECT ?annotation ?title ?term ?length1 ?length
  WHERE 
     {?annotationZ wdt:P19 ?anchorZ.
      ?anchorZ wdt:P7 wd:Q6.
      ?anchorZ wdt:P12 ?article.
      ?anchorZ wdt:P10 ?length.
      ?article wdt:P20 ?title.
      ?annotation wdt:P19 ?anchor.
      ?anchor wdt:P12 ?article.
      ?annotation wdt:P15 ?term.
      ?anchor wdt:P10 ?length1.
     FILTER(10*?length1 < ?length)
     }
  ORDER BY ASC (?length1)
  

  Try it!

LineChart by decile for given dictionary and paper

• #defaultView:LineChart
  SELECT ?decile (COUNT(?annotation) AS ?count)
  WHERE
  {
      ?annotation wdt:P3 wd:Q5 ;
                  wdt:P19 ?anchor;
                  wdt:P16 "infectiousdiseasesdrugs".
      ?anchor wdt:P12 wd:Q6679.
      ?anchor wdt:P10 ?charnumber.
      ?annotationZ wdt:P19 ?anchorZ.
      ?anchorZ wdt:P7 wd:Q6;
               wdt:P12 wd:Q6679;
               wdt:P10 ?length.
     BIND (floor(10*?charnumber/?length) AS ?decile)
    
    SERVICE wikibase:label { bd:serviceParam wikibase:language "[AUTO_LANGUAGE],en". }
    }
   GROUP BY ?decile
  

  Try it!