Examples of ProcessingDocument

uk.ac.cam.ch.wwmm.oscar3.recogniser.document.ProcessingDocument
A document, with data structures to store information such as tokens. This extra information is essential for many document processing tasks. These should be created using the ProcessingDocumentFactory class. @author ptc24

Examples of uk.ac.cam.ch.wwmm.oscar3.recogniser.document.ProcessingDocument

    
    //NameRecogniser nr = new NameRecogniser();
    //nr.halfProcess(sourceDoc);
    //nr.makeTokenisers(false);


    ProcessingDocument procDoc = ProcessingDocumentFactory.getInstance().makeTokenisedDocument(sourceDoc, false, false, false, safDoc);
    
    Nodes n = safDoc.query("/saf/annot[slot[@name='type']['PRW']]");


    Map<TokenSequence,Boolean> tokSeqs = new HashMap<TokenSequence,Boolean>();
    Map<TokenSequence,Bag<String>> tokSeqPRWs = new HashMap<TokenSequence,Bag<String>>();
    
    for(int i=0;i<n.size();i++) {
      Element e = (Element)n.get(i);
      Token token = procDoc.getTokenByStart(e.getAttributeValue("from"));
      if(token == null) token = procDoc.getTokenByEnd(e.getAttributeValue("to"));
      if(token == null) {
        System.out.println("Eeep!");
      } else {
        TokenSequence tokSeq = token.getTokenSequence();
        //boolean isReact = "REACT".equals(SafTools.getSlotValue(e, "subtype"));

View Full Code Here

Examples of uk.ac.cam.ch.wwmm.oscar3.recogniser.document.ProcessingDocument

    ResourceGetter rg = new ResourceGetter("uk/ac/cam/ch/wwmm/oscar3/test/testcard/resources/");
    String s = rg.getString("testcard.txt");
    assertTrue("Have testcard string", s != null && s.length() > 0);
    Document doc = TextToSciXML.textToSciXML(s);


    ProcessingDocument procDoc = ProcessingDocumentFactory.getInstance().makeTokenisedDocument(doc, false, false, false);
    assertTrue(procDoc != null);
    List<NamedEntity> neList;
    ChemicalEntityRecogniser cei = new PatternRecogniser();
    neList = cei.findNamedEntities(procDoc);
    assertTrue(neList != null);

View Full Code Here

Examples of uk.ac.cam.ch.wwmm.oscar3.recogniser.document.ProcessingDocument

    ResourceGetter rg = new ResourceGetter("uk/ac/cam/ch/wwmm/oscar3/test/testcard/resources/");
    String s = rg.getString("testcard.txt");
    assertTrue("Have testcard string", s != null && s.length() > 0);
    Document doc = TextToSciXML.textToSciXML(s);
    
    ProcessingDocument procDoc = ProcessingDocumentFactory.getInstance().makeTokenisedDocument(doc, false, false, false);
    assertTrue(procDoc != null);
    List<NamedEntity> neList;
    ChemicalEntityRecogniser cei = new MEMMRecogniser();
    neList = cei.findNamedEntities(procDoc);
    assertTrue(neList != null);

View Full Code Here

Examples of uk.ac.cam.ch.wwmm.oscar3.recogniser.document.ProcessingDocument

            Element e = (Element)n.get(i);
            e.addAttribute(new Attribute("type", "CHEMICAL"));
            //XOMTools.removeElementPreservingText((Element)n.get(i));
          }          
        }
        ProcessingDocument procDoc = ProcessingDocumentFactory.getInstance().makeTokenisedDocument(doc, true, false, false);


        //NameRecogniser nr = new NameRecogniser();
        //nr.halfProcess(doc);


        //if(memm.patternFeatures) {
        //  nr.findForReps(cheatTokenisation);          
        //} else {
          //nr.makeTokenisers(cheatTokenisation);
        //}
        
        int paperTestNEs = 0;
        
        List<Double> paperGoodProbs = new ArrayList<Double>();
        List<Double> paperBadProbs = new ArrayList<Double>();


        Map<NamedEntity,Double> confidences = new HashMap<NamedEntity,Double>();
        
        List<NamedEntity> entities;
        
        Set<String> testNEs = new LinkedHashSet<String>();


        for(TokenSequence tokSeq : procDoc.getTokenSequences()) {
          Nodes neNodes = tokSeq.getElem().query(".//ne");
          for(int k=0;k<neNodes.size();k++) {
            Element neElem = (Element)neNodes.get(k);
            if(filterType != null && !neElem.getAttributeValue("type").equals(filterType)) continue;
            if(antiFilterType != null && neElem.getAttributeValue("type").equals(antiFilterType)) continue;

View Full Code Here

Examples of uk.ac.cam.ch.wwmm.oscar3.recogniser.document.ProcessingDocument

      }*/
      //NameRecogniser nr = new NameRecogniser();
      //nr.halfProcess(xmlDoc);
      //nr.makeTokenisers(false);
      
      ProcessingDocument procDoc = ProcessingDocumentFactory.getInstance().makeTokenisedDocument(xmlDoc, false, false, false);
      
      //for(Tokeniser t : nr.getTokenisers()) {
        TokenStream ts = new Oscar3TokenStream(procDoc.getTokenSequences());
        ts = new Oscar3Filters(ts);
        luceneDoc.add(new Field("txt", ts, Field.TermVector.WITH_POSITIONS_OFFSETS));
        
      //}

View Full Code Here

Examples of uk.ac.cam.ch.wwmm.oscar3.recogniser.document.ProcessingDocument

   * @param sourceDoc The SciXML document to parse, now unlikely to be altered.
   * @param safDoc An (empty) SAF XML document.
   * @throws Exception C
   */
  private void processDocumentInternal(Document sourceDoc, Document safDoc) throws Exception {
    ProcessingDocument procDoc = ProcessingDocumentFactory.getInstance().makeTokenisedDocument(new Document((Element)XOMTools.safeCopy(sourceDoc.getRootElement())), false, false, false);
    List<NamedEntity> neList;
    ChemicalEntityRecogniser cer = null;
    try
    {
      cer = (ChemicalEntityRecogniser) Class.forName(Oscar3Props.getInstance().chemicalEntityRecogniser).newInstance();

View Full Code Here

0 1 2

TOP

All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.