Package uk.ac.cam.ch.wwmm.oscar3.recogniser.document

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

58
59
60
61
62
63
64
65
66
67
68
69
   
    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);
        wordCounts.add(word);
      }
    } 
View Full Code Here
96
97
98
99
100
101
102
103
104
105
106
107
108
    Collections.sort(offsetArray, offsetComparator);
    System.out.println(System.currentTimeMillis() - time); 
  }
 
  public int [] searchForString(String searchString) {
    TokenSequence ts = Tokeniser.getInstance().tokenise(searchString);
    List<Integer> tsl = new ArrayList<Integer>(ts.size() + 1);
    for(Token t : ts.getTokens()) {
      String s = t.getValue().toLowerCase();
      if(tokenIndex.containsKey(s)) {
        tsl.add(tokenIndex.get(s));
      } else {
        return null;
View Full Code Here
134
135
136
137
138
139
140
141
142
143
144
      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"));
        if(tokSeqs.containsKey(tokSeq)) {
          if(isReact) tokSeqs.put(tokSeq, true);
        } else {
          tokSeqs.put(tokSeq, isReact);
View Full Code Here
26
27
28
29
30
31
32
33
34
35
36
        c = reader.read();
      }     
    } catch (Exception e) {
      throw new Error(e);
    }
    TokenSequence t = Tokeniser.getInstance().tokenise(sb.toString());
   
    TokenStream ts = new Oscar3TokenStream(t);
    ts = new Oscar3Filters(ts);
    return ts;
  }
View Full Code Here
134
135
136
137
138
139
140
141
142
143
144
      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
67
68
69
70
71
72
73
74
75
76
77
    //files = files.subList(0, 10);
   
   
    Bag<String> tokenBag = new Bag<String>();
   
    TokenSequenceSource tss = new TokenSequenceSource(files);
    int i=0;
    for(TokenSequence ts : tss) {
      for(Token t : ts.getTokens()) {
        tokenBag.add(t.getValue().intern());
      }
View Full Code Here
45
46
47
48
49
50
51
52
53
54
55
      corpusOffset++;
    }
  }
 
  public InverseSearcher(List<File> files) throws Exception {
    TokenSequenceSource tss = new TokenSequenceSource(files);

    corpusArray = new ArrayList<Integer>();
    offsetArray = new ArrayList<Integer>();
    tokenList = new ArrayList<String>();
    tokenIndex = new HashMap<String,Integer>();
View Full Code Here
22
23
24
25
26
27
28
29
30
31
32
    }
    Bag<String> allTerms = new Bag<String>();
    String word = "reduction";
    List<File> files = FileTools.getFilesFromDirectoryByName(new File("/home/ptc24/newows/corpora/BioIE"), "source.xml");
    //List<File> files = FileTools.getFilesFromDirectoryByName(new File("/scratch/pubmed/2005"), "source.xml");
    TokenSequenceSource ts = new TokenSequenceSource(files);
    for(TokenSequence t : ts) {
      List<String> tokens = t.getTokenStringList();
      for(int i=0;i<tokens.size();i++) {
        if(tokens.get(i).equals(word)) {
          for(int j=Math.max(0, i-windowSize);j<=Math.min(i+windowSize,tokens.size()-1);j++) {
View Full Code Here
46
47
48
49
50
51
52
53
54
55
56
57
      for(String word : t.getTokenStringList()) {
        terms.add(StringTools.normaliseName(word));
      }
    }*/
   
    TokenSequenceSource tokSeqSource = new TokenSequenceSource(files);
    tokSeqSource.reset();
    for(TokenSequence t : tokSeqSource) {
      for(String word : t.getTokenStringList()) {
        word = StringTools.normaliseName(word);
        word = word.replaceAll("\\s+", "_");
        terms.add(word.intern());
View Full Code Here
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
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
124
125
126
127
128
129
      beforeWidth = allowableWidth / 3;
      afterWidth = (2 * allowableWidth) / 3;     
    }*/
   
    //StringSource ss = new StringSource(files, false);
    TokenSequenceSource ts = new TokenSequenceSource(files);
    boolean doSort = true;
    for(TokenSequence t : ts) {
      //s = s.replaceAll("\\s+", " ");
      //Tokeniser t = new Tokeniser(null);
      //t.tokenise(s);
      int sstart = t.getOffset();
      int send = t.getSourceString().length() + sstart;
      for(Token token : t.getTokens()) {
        if(token.getValue().equalsIgnoreCase(word)) {
          int wstart = token.getStart();
          int wend = token.getEnd();
          String before = t.getStringAtOffsets(Math.max(sstart, wstart-beforeWidth), wstart);
          before = before.replaceAll("\\s+", " ");
          if(before.startsWith(" ")) before = "." + before.substring(1);
          if(before.length() < beforeWidth) {
            before = StringTools.multiplyString(".", beforeWidth - before.length()) + before;
          }
          //if(wstart < beforeWidth) {
          //  before = StringTools.multiplyString(" ", beforeWidth - wstart) + t.getStringAtOffsets(0, wstart);
          //} else {
          //  before = t.getStringAtOffsets(wstart - beforeWidth, wstart);
          //}
          String after = t.getStringAtOffsets(wend, Math.min(send, wend + afterWidth));
          after = after.replaceAll("\\s+", " ");
          //if(s.length() - wend > afterWidth) {
          //  after = s.substring(wend, wend + afterWidth);
          //} else {
          //  after = s.substring(wend);
          //}
          String display = before + "  " + token.getValue() + "  " + after;
          String sort = "";
          if("interleave".equals(mode)) {
            sort = interleave(before, after, true);           
          } else if("left".equals(mode)) {
            StringBuffer sb = new StringBuffer(before.toLowerCase());
            sb.reverse();
            sort = sb + "\n" + after.toLowerCase();
          } else if("right".equals(mode)) {
            sort = after.toLowerCase() + "\n" + before.toLowerCase();           
          } else {
            doSort = false;
          }
          ConcordanceEntry entry = new ConcordanceEntry();
          entry.text = display;
          entry.start = token.getStartXPoint();
          entry.end = token.getEndXPoint();
          entry.file = ts.getCurrentFile().getAbsolutePath();
          concordance.put(entry, sort);
        }
      }
    }
    if(!doSort) {
View Full Code Here
TOP

Related Classes of uk.ac.cam.ch.wwmm.oscar3.recogniser.document.TokenSequenceSource

  • 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.