2023 stuff for CiTO annotations, related to https://scholia.toolforge.org/work/Q117357537

README.md

2023 stuff for CiTO annotations

Some files related to some data curation I did yesterday to add some 200~ish CiTO annotations to Wikidata. The data is CCZero, from https://doi.org/10.5281/zenodo.6885109

There are two files here:

• cits_cleaned.csv derived from the cits.csv from the Zenodo data
• quickstatements.groovy that takes cits_cleaned.csv as input and creates QuickStatements

cits_cleaned.csv

quickstatements.groovy

// Copyright (C) 2023  Egon Willighagen
// License: MIT

@Grab(group='org.apache.commons', module='commons-csv', version='1.10.0')

// Bacting config
@Grab(group='io.github.egonw.bacting', module='managers-ui', version='0.3.1')
@Grab(group='io.github.egonw.bacting', module='managers-rdf', version='0.3.1')

import org.apache.commons.csv.*
import java.text.SimpleDateFormat;
import java.util.Date;

workspaceRoot = ".."
ui = new net.bioclipse.managers.UIManager(workspaceRoot);
bioclipse = new net.bioclipse.managers.BioclipseManager(workspaceRoot);
rdf = new net.bioclipse.managers.RDFManager(workspaceRoot);

DOIs = new java.util.HashSet()

// parse a first time: get the DOIs
csvData = new FileReader(new File("cits3.csv"))
records = CSVFormat.RFC4180.withFirstRecordAsHeader().parse(csvData); 
for (CSVRecord csvRecord : records) {
  // println "Cited: " + csvRecord.get("cited_doi").trim().toUpperCase()
  if (csvRecord.isSet("cited_doi") && csvRecord.isSet("citing_doi")) {
    cited = csvRecord.get("cited_doi").trim().toUpperCase()
    citing = csvRecord.get("citing_doi").trim().toUpperCase()
    if (!DOIs.contains(cited)) { DOIs.add(cited) }
    if (!DOIs.contains(citing)) { DOIs.add(citing) }
  }
}

// println "Found DOIs: " + (DOIs.size())

// lookup the Wikidata IDs
values = ""
DOIs.each { doi ->
  values += "\"${doi.toUpperCase()}\" \n"
}
sparql = "SELECT DISTINCT ?work ?doi WHERE {\n VALUES ?doi {\n ${values} }\n ?work wdt:P356 ?doi .\n}"
if (bioclipse.isOnline()) {
  rawResults = bioclipse.sparqlRemote("https://query.wikidata.org/sparql", sparql  )
  results = rdf.processSPARQLXML(rawResults, sparql)
}
// make a map
map = new HashMap<String,String>()
for (i=1;i<=results.rowCount;i++) {
  rowVals = results.getRow(i)
  map.put(rowVals[1], rowVals[0].replace("http://www.wikidata.org/entity/",""))
}

intentionTypes = new HashMap<String,String>()
intentionTypes.put("obtains background from","Q96480394")
intentionTypes.put("obtains support from", "Q115470993")
intentionTypes.put("cites for information", "Q96479970")
intentionTypes.put("discusses", "Q96471822")
intentionTypes.put("uses method in", "Q96472102")
intentionTypes.put("describes", "Q117358004")
intentionTypes.put("qualifies", "Q117121930")
intentionTypes.put("cites as evidence", "Q117358022")
intentionTypes.put("disputes", "Q117121923")
intentionTypes.put("extends", "Q96472100")
intentionTypes.put("parodies", "Q117121928")
intentionTypes.put("cites as source document", "Q111736247")
intentionTypes.put("speculates on", "Q117121933")
intentionTypes.put("critiques", "Q105624924")
intentionTypes.put("agrees with", "Q96604140")
intentionTypes.put("uses conclusions from", "Q117121934")

// parse a second time: create QuickStatements
csvData = new FileReader(new File("cits3.csv"))
records = CSVFormat.RFC4180.withFirstRecordAsHeader().parse(csvData);
date = new SimpleDateFormat("yyyy-MM-dd").format(new Date());
println "qid,P2860,qal3712,S248,s813"
for (CSVRecord csvRecord : records) {
  println "# Row: ${csvRecord.toString()}"
  if (csvRecord.isSet("cited_doi") && csvRecord.isSet("citing_doi")) {
    cited = csvRecord.get("cited_doi").trim().toUpperCase()
    citing = csvRecord.get("citing_doi").trim().toUpperCase()
    intentions = csvRecord.get("intext_citation.intent").toLowerCase().split(";;")
    for (intention : intentions) {
      intention = intention.trim()
      qid = map.get(citing); // if (qid == null) println "# DOI not in Wikidata: ${citing}"
      citedQID = map.get(cited); // if (citedQID == null) println "# DOI not in Wikidata: ${cited}"
      done = new HashSet<String>()
      if (qid != null && citedQID != null) {
        if (intentionTypes.containsKey(intention)) {
          intentionQID = intentionTypes.get(intention)
          if (!(done.contains(intentionQID))) {
            println "${qid},${citedQID},${intentionQID},Q117357537,+${date}T00:00:00Z/11"
            done.add(intentionQID)
          }
        } else {
          println "# Unrecognized intention: ${intention}"
        }
      }
    }
  }
}