Package org.apache.ctakes.typesystem.type.syntax

Examples of org.apache.ctakes.typesystem.type.syntax.BaseToken

162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
        }
      }

      // Initialize Token / Sentence info for the ClearParser Semantic Role Labeler
      for (int i = 0; i < tokens.size(); i++) {
        BaseToken token = tokens.get(i);

        // Determine HeadId
        DepNode node = new DepNode();
        ConllDependencyNode casDepNode = JCasUtil.selectCovered(jCas, ConllDependencyNode.class, token).get(0);
        casDepNode.getDeprel();
        String headRelation = casDepNode.getDeprel();
        ConllDependencyNode head = casDepNode.getHead();
       
        // If there is no head, this is the head node, set node to 0
        int headId = (head == null) ? 0 : depNodeToID.get(head);

        // Populate Dependency Node / Tree information
        node.id = i + 1;
        node.form = token.getCoveredText();
        node.pos = token.getPartOfSpeech();
        node.lemma = useLemmatizer ? "" : token.getNormalizedForm();
        node.setHead(headId, headRelation, 0);
        tree.add(node);
      }
      tree.setPredicates(AbstractReader.LANG_EN);
View Full Code Here
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
    // Start at node 1, since node 0 is considered the head of the sentence
    for (int i = 1; i < tree.size(); i++) {
      // Every ClearParser parserNode will contain an srlInfo field.
      DepNode parserNode = tree.get(i);
      BaseToken token = tokens.get(i - 1);
      if (parserNode.srlInfo == null) {
        continue;
      }

      if (parserNode.srlInfo.isPredicate()) {
        int headId = i;
        if (!headIdToPredicate.containsKey(headId)) {
          // We have not encountered this predicate yet, so create it
          Predicate pred = this.createPredicate(jCas, parserNode.srlInfo.rolesetId, token);
          headIdToPredicate.put(headId, pred);
          pred.setRelations(new EmptyFSList(jCas));
        }
      } else {
        for (SRLHead head : parserNode.srlInfo.heads) {
          Predicate predicate;

          // Determine which predicate this argument belongs to
          if (!headIdToPredicate.containsKey(head.headId)) {
            // The predicate hasn't been encountered, so create it
            BaseToken headToken = tokens.get(head.headId - 1);
            predicate = this.createPredicate(jCas, parserNode.srlInfo.rolesetId, headToken);
            headIdToPredicate.put(head.headId, predicate);
          } else {
            predicate = headIdToPredicate.get(head.headId);
          }
View Full Code Here
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
130
131
132
133
134
135
136
137
138
139
140
141
           
           
            ListIterator<BaseToken>           itt = tokens.listIterator();
            ListIterator<ConllDependencyNode> itn = nodes.listIterator();
            BaseToken           bt = null;
            ConllDependencyNode dn = null;
            if (tokens.size()>0 && nodes.size()>0) {
                // iterate through the parallel sorted lists
                if (itt.hasNext()) bt                  = itt.next();
                if (itn.hasNext()) dn                  = itn.next();
                if (dn != null)
                    if (dn.getId()==0 && itn.hasNext())
                        dn = itn.next();
                while (itt.hasNext() || itn.hasNext()) {
                    if (bt.getBegin()==dn.getBegin() ) { // Allow ragged right edge //&& bt.getEnd()==dn.getEnd()) {
                        dn.setLemma( bt.getNormalizedForm() );
                        if (dn.getLemma()==null)
                            dn.setLemma( dn.getForm().toLowerCase() );
                        dn.addToIndexes();
                        if (itt.hasNext()) bt = itt.next();
                        if (itn.hasNext()) dn = itn.next();
                    } else if ( bt.getBegin()<dn.getBegin() ) {
                        if (itt.hasNext()) bt = itt.next(); else break;
                    } else if ( bt.getBegin()>dn.getBegin() ) {
                        // not every node will get a lemma b/c not all tokens are word tokens
                        if (itn.hasNext()) {
                            dn.setLemma( dn.getForm().toLowerCase() );
                            dn.addToIndexes();
                            dn = itn.next();
                        } else break;
                    }
                }
                if (bt.getBegin()==dn.getBegin() && bt.getEnd()==dn.getEnd()) {
                    dn.setLemma( bt.getNormalizedForm() );
                    dn.addToIndexes();
                }
            }
           
    }
View Full Code Here
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
      List<BaseToken> tokens = JCasUtil.selectCovered(jCas, BaseToken.class, sentence);
      DepTree tree = new DepTree();

      // Convert CAS data into structures usable by ClearParser
      for (int i = 0; i < tokens.size(); i++) {
        BaseToken token = tokens.get(i);
        DepNode node = new DepNode();
        node.id = i + 1;
        node.form = token.getCoveredText();
        node.pos = token.getPartOfSpeech();
        node.lemma = useLemmatizer ? lemmatizer.getLemma(node.form, node.pos) : token.getNormalizedForm();
        tree.add(node);
      }

      // Run parser and convert output back to CAS friendly data types
      parser.parse(tree);
View Full Code Here
48
49
50
51
52
53
54
55
56
57
    // index the base tokens and NEs by their offsets
    hbs = new Hashtable<Integer, BaseToken>();
    hbe = new Hashtable<Integer, BaseToken>();
    FSIterator iter = jcas.getJFSIndexRepository().getAnnotationIndex(BaseToken.type).iterator();
    while (iter.hasNext()) {
      BaseToken t = (BaseToken) iter.next();
      hbs.put(t.getBegin(), t);
      hbe.put(t.getEnd(), t);
    }
  }
View Full Code Here
60
61
62
63
64
65
66
67
68
69
70
    stopwords = l;
  }

  static boolean isPronoun (Markable m) {
    if (m.getContent() instanceof BaseToken) {
      BaseToken t = (BaseToken) m.getContent();
      if (t.getPartOfSpeech().startsWith("PRP")) // TODO: since only 3rd person pronouns are added as markables, no need to check
        return true;
    }
    return false;
  }
View Full Code Here
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
    return "U";
  }

  public ArrayList<BaseToken> containedTokens (int a, int b) {
    ArrayList<BaseToken> ret = new ArrayList<BaseToken>();
    BaseToken t1 = hbs.get(a);
    BaseToken t2 = hbe.get(b);
    if (t1!=null && t2!=null) {
      int begin = t1.getTokenNumber();
      int end = t2.getTokenNumber();
      LinkedList<Annotation> l = FSIteratorToList.convert(jcas.getJFSIndexRepository().getAnnotationIndex(BaseToken.type).iterator());
      for (int i = 0; i < l.size(); i++) {
        BaseToken t = (BaseToken) l.get(i);
        if (t.getTokenNumber()>=begin && t.getTokenNumber()<=end)
          ret.add(t);
      }
    }
//    int e;
//    while (t!=null && (e=t.getEnd())<=b) {
View Full Code Here
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
130
131
132
133
                }
            }

            ListIterator<BaseToken> itt           = tokens.listIterator();
            ListIterator<ConllDependencyNode> itn = nodes.listIterator();
            BaseToken           bt = null;
            ConllDependencyNode dn = null;
            if (tokens.size()>0 && nodes.size()>0) {
                // iterate through the parallel sorted lists
              if (itt.hasNext()) bt                  = itt.next();
              if (itn.hasNext()) dn                  = itn.next();
              if (dn != null)
                if (dn.getId()==0 && itn.hasNext())
                  dn = itn.next();
              while (itt.hasNext() || itn.hasNext()) {
                if (bt.getBegin()==dn.getBegin() && bt.getEnd()==dn.getEnd()) {
                  dn.setPostag( bt.getPartOfSpeech() );
                  dn.setCpostag( bt.getPartOfSpeech() ); 
                  dn.addToIndexes();
                  if (itt.hasNext()) bt = itt.next();
                  if (itn.hasNext()) dn = itn.next();
                } else if ( bt.getBegin()<dn.getBegin() ) {
                  if (itt.hasNext()) bt = itt.next(); else break;
                } else if ( bt.getBegin()>dn.getBegin() ) {
                  if (itn.hasNext()) dn = itn.next(); else break;
                }
              }
              if (bt.getBegin()==dn.getBegin() && bt.getEnd()==dn.getEnd()) {
                dn.setPostag( bt.getPartOfSpeech() );
                dn.setCpostag( bt.getPartOfSpeech() ); 
                dn.addToIndexes();
              }
            }
             
    }
View Full Code Here
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
    JFSIndexRepository indexes = jcas.getJFSIndexRepository();
    Iterator btaItr = indexes.getAnnotationIndex(BaseToken.type).iterator();
    while (btaItr.hasNext())
    {
      BaseToken bta = (BaseToken) btaItr.next();
      if (!((bta instanceof NewlineToken)
          || (bta instanceof PunctuationToken)
          || (bta instanceof ContractionToken)
          || (bta instanceof SymbolToken)))
      {
        LookupToken lt = new LookupAnnotationToJCasAdapter(bta);

        // POS exclusion logic for first word lookup
        if (isTagExcluded(bta.getPartOfSpeech()))
        {
          lt.addStringAttribute(
              FirstTokenPermutationImpl.LT_KEY_USE_FOR_LOOKUP,
              "false");
        }
View Full Code Here
75
76
77
78
79
80
81
82
83
84
85
    JFSIndexRepository indexes = jcas.getJFSIndexRepository();
    Iterator btaItr = indexes.getAnnotationIndex(BaseToken.type)
        .iterator();
    while (btaItr.hasNext())
    {
      BaseToken bta = (BaseToken) btaItr.next();
      if (!((bta instanceof NewlineToken)
          || (bta instanceof PunctuationToken)
          || (bta instanceof ContractionToken)
          || (bta instanceof SymbolToken)))
      {
View Full Code Here
TOP

Related Classes of org.apache.ctakes.typesystem.type.syntax.BaseToken

  • org.apache.ctakes.assertion.eval.AssertionEvalBasedOnModifier$ReferenceSupportingAnnotationsSystemToGoldCopier
  • org.apache.ctakes.assertion.eval.AssertionEvaluation$ReferenceSupportingAnnotationsSystemToGoldCopier
  • org.apache.ctakes.assertion.medfacts.cleartk.AlternateCuePhraseAnnotator
  • org.apache.ctakes.assertion.medfacts.cleartk.extractors.SRLFeatureExtractor
  • org.apache.ctakes.assertion.medfacts.i2b2.api.CharacterOffsetToLineTokenConverterCtakesImpl
  • org.apache.ctakes.assertion.medfacts.i2b2.api.SingleDocumentProcessorCtakes
  • org.apache.ctakes.chunker.ae.Chunker
  • org.apache.ctakes.clinicalpipeline.ae.ExtractionPrepAnnotator
  • org.apache.ctakes.constituency.parser.MaxentParserWrapper
  • org.apache.ctakes.constituency.parser.util.TreeUtils
    • 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.