Package uk.ac.cam.ch.wwmm.oscar3.indexersearcher

Examples of uk.ac.cam.ch.wwmm.oscar3.indexersearcher.SnippetGetter$SnippetComparator

494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
    }
   
    //if(surface.matches("([Pp]oly).+")) features.add("polymer");
    //if(surface.matches(".+\\(\\d\\d\\d+\\)")) features.add("surfacenotation");
   
    Token t = state.procDoc.getTokenByStart(annot.getAttributeValue("from"));
    if(fPrevious && t != null) {
      Token tt = t.getNAfter(-1);
      if(tt != null) {
        if(TokenTypes.isRef(tt) && tt.getNAfter(-1) != null) {
          //  features.add(prefix + "skiprefprev");
          tt = tt.getNAfter(-1);
        }
        String ttv = tt.getValue();
        ttv = ttv.replaceAll("\\s+", "_");
        features.add("prev=" + ttv);
      }
    }
   
    t = state.procDoc.getTokenByEnd(annot.getAttributeValue("to"));
    if(fNext && t != null) {
      Token tt = t.getNAfter(1);
      if(tt != null && tt.getValue().equals("-")) {
        tt = t.getNAfter(2);
      }
      if(tt != null) {
        if(TokenTypes.isRef(tt) && tt.getNAfter(1) != null) {
          tt = tt.getNAfter(1);
        }
       
        String ttv = tt.getValue();
        ttv = ttv.replaceAll("\\s+", "_");
        features.add("next=" + ttv);
      }
    }
   
View Full Code Here
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
    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"));
        boolean isPubmed = f.toString().contains("pubmed");
        if(tokSeqs.containsKey(tokSeq)) {
          if(isPubmed) tokSeqs.put(tokSeq, true);
        } else {
View Full Code Here
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
      }
      prevCell = cell;
    }
    //System.out.println("**************************");
    // Parenthetical phrases
    Token beforeBracket = null;
    Token prevToken = null;
    for(Token token : tokSeq.getTokens()) {
      if(prevToken == null) {
        prevToken = token;
        continue;
      }
      if(beforeBracket == null) {
        if(token.getValue().equals("(")) beforeBracket = prevToken;
      } else if(token.getValue().equals(")")) {
        LatticeCell beforeBracketCell = tokensToTokenCells.get(beforeBracket);
        LatticeCell bracketCell = tokensToTokenCells.get(token);
        if(beforeBracketCell != null || bracketCell != null) {
          beforeBracketCell.addInheritance(bracketCell);
          //System.out.println("Parenthetical: " + tokSeq.getSubstring(beforeBracket.getId(), token.getId()));
        }
        beforeBracket = null;
      }
      prevToken = token;
    }
   
    // Build in named entities
    for(Token token : tokSeq.getTokens()) {
      if(startsToNEs.containsKey(token.getStart())) {
        for(Element neElem : startsToNEs.get(token.getStart())) {
          Token endToken = procDoc.getTokenByEnd(neElem.getAttributeValue("to"));
          if(endToken == null) continue;
          prevCell = tokenToPrev.get(token);
          LatticeCell endTokenCell = tokensToTokenCells.get(endToken);
          if(endTokenCell == null) continue;
          LatticeCell neCell = new LatticeCell(neElem, endTokenCell);
View Full Code Here
104
105
106
107
108
109
110
111
   *
   * @param word The string to test.
   */
  public static void destroyInstanceIfWordTokenises(String word) {
    if(myInstance == null) return;
    TokenSequence ts = Tokeniser.getInstance().tokenise(word);
    if(ts.getTokens().size() > 1) myInstance = null;
  }
View Full Code Here
127
128
129
130
131
132
133
134
   *
   * @param word The string to test.
   */
  public static void destroyInstanceIfWordTokenises(String word) {
    if(myInstance == null) return;
    TokenSequence ts = Tokeniser.getInstance().tokenise(word);
    if(ts.getTokens().size() > 1) myInstance = null;
  }
View Full Code Here
133
134
135
136
137
138
139
140
141
142
143
144
        ontDFANumber++;
      }
      type = type + "_" + ontDFANumber;*/
    }

    TokenSequence ts = Tokeniser.getInstance().tokenise(ne);
    List<String> tokens = ts.getTokenStringList();

    if(!alwaysAdd && tokens.size() == 1 && !ne.contains("$")) return;
    StringBuffer sb = new StringBuffer();
    for(String token : tokens) {
      sb.append(getRepForToken(token));
View Full Code Here
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
                featuresForAbbrev.add(abbrMode + "abbr<" + thresh);
              }             
            }
          } else {
            int tokID = ne.getFirstToken().getId();
            TokenSequence tokSeq = ne.getFirstToken().getTokenSequence();
            int length = surf.length();
            boolean isAcro = false;
            if(allCaps.matcher(surf).matches()) {
              if(length <= (tokID - 1)) {
                isAcro = true;
                for(int i=0;i<length;i++) {
                  if(!tokSeq.getToken(tokID - length - 1 + i).getValue().toUpperCase().startsWith(surf.substring(i,i+1))) isAcro = false;
                }
                if(isAcro) {
                  featuresForAbbrev.add("allUpperAbbrev");
                }
              }
View Full Code Here
126
127
128
129
130
131
132
133
134
135
136
    double conf = ne.getConfidence();
    double confLog = Math.log(conf) - Math.log(1 - conf);

    List<String> features = new ArrayList<String>();
   
    TokenSequence t = ne.getTokens().get(0).getTokenSequence();
    int entityLength = ne.getTokens().size();
    int startID = ne.getTokens().get(0).getId();
    int endID = startID + entityLength - 1;
    String surf = ne.getSurface();
   
View Full Code Here
57
58
59
60
61
62
63
64
65
66
67
68
     
      Bag<String> wordCounts = new Bag<String>();
     
      ss.reset();
      for(String s : ss) {
        TokenSequence t = Tokeniser.getInstance().tokenise(s);
        for(String word : t.getTokenStringList()) {
          if(!word.matches(".*[a-z][a-z].*")) continue;
          word = StringTools.normaliseName(word);
          if(!knownWords.contains(NGram.parseWord(word))) wordCounts.add(word);
        }
      } 
View Full Code Here
52
53
54
55
56
57
58
59
60
61
62
63
64
65
      Elements lineElems = elems.get(i).getChildElements();
      //List<String> wordsForWord = new ArrayList<String>();
      Set<String> wordsForWord = new HashSet<String>();
      for(int j=0;j<lineElems.size();j++) {
        String lineVal = lineElems.get(j).getValue();
        TokenSequence t = Tokeniser.getInstance().tokenise(lineVal);
        boolean found = false;
        List<String> tl = new ArrayList<String>();
        for(String s : t.getTokenStringList()) {
          if(s.compareToIgnoreCase(name) == 0) {
            tl.add("*WORD*");
            found = true;
          } else {
            //tl.add(StringTools.normaliseName2(s));
View Full Code Here
TOP

Related Classes of uk.ac.cam.ch.wwmm.oscar3.indexersearcher.SnippetGetter$SnippetComparator

  • dk.brics.automaton.Automaton
  • jregex.Pattern
  • nu.xom.Attribute
  • nu.xom.Builder
  • nu.xom.Document
  • nu.xom.Element
  • nu.xom.Elements
  • nu.xom.Node
  • nu.xom.Nodes
  • nu.xom.Serializer
    • Copyright © 2018 www.massapicom. All rights reserved.
    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.