Package org.cleartk.classifier.feature.extractor

Examples of org.cleartk.classifier.feature.extractor.CleartkExtractor

183
184
185
186
187
188
189
190
191
192
193
194
195
        charExtractors,
        new TypePathExtractor(BaseToken.class, "partOfSpeech"));
   
    featureExtractors = new ArrayList<SimpleFeatureExtractor>();
//    featureExtractors.add(new CleartkExtractor(BaseToken.class, new CoveredTextExtractor(), new Bag(new Covered())));
    featureExtractors.add(new CleartkExtractor(BaseToken.class, allExtractors, new Bag(new Covered())));
//    featureExtractors.add(charExtractors);
    wordTypeExtractor = new CleartkExtractor(BaseToken.class, new TimeWordTypeExtractor(), new Bag(new Covered()));
//    featureExtractors.add(new CleartkExtractor(BaseToken.class, new CoveredTextExtractor(), new Bag(new Preceding(1))));
//   featureExtractors.add(new CleartkExtractor(BaseToken.class, new CoveredTextExtractor(), new Bag(new Following(1))));
    // bag of constituent descendent labels
//    featureExtractors.add(new CleartkExtractor(TreebankNode.class, new TypePathExtractor(TreebankNode.class, "nodeType"), new Bag(new Covered())));
   
View Full Code Here
221
222
223
224
225
226
227
228
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
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
    // a list of feature extractors that require the token and the sentence
//    this.contextFeatureExtractors = new ArrayList<CleartkExtractor>();
   
    this.tokenCleartkExtractors = new ArrayList<CleartkExtractor>();

    CleartkExtractor tokenExtraction1 =
        new CleartkExtractor(
            BaseToken.class,
//            new FeatureFunctionExtractor(new CoveredTextExtractor(), new LowerCaseFeatureFunction()),
//            new FeatureFunctionExtractor(new CoveredTextExtractor(), new BrownClusterFeatureFunction()),
            new CoveredTextExtractor(),
            //new CleartkExtractor.Covered(),
            new CleartkExtractor.LastCovered(2),
            new CleartkExtractor.Preceding(5),
            new CleartkExtractor.Following(4),
            new CleartkExtractor.Bag(new CleartkExtractor.Preceding(3)),
            new CleartkExtractor.Bag(new CleartkExtractor.Following(3)),
            new CleartkExtractor.Bag(new CleartkExtractor.Preceding(5)),
            new CleartkExtractor.Bag(new CleartkExtractor.Following(5)),
            new CleartkExtractor.Bag(new CleartkExtractor.Preceding(10)),
            new CleartkExtractor.Bag(new CleartkExtractor.Following(10))
            );
   
//    CleartkExtractor posExtraction1 =
//        new CleartkExtractor(
//            BaseToken.class,
//            new TypePathExtractor(BaseToken.class, "partOfSpeech"),
//            new CleartkExtractor.LastCovered(2),
//            new CleartkExtractor.Preceding(3),
//            new CleartkExtractor.Following(2)
//            );

    this.tokenCleartkExtractors.add(tokenExtraction1);
    //this.tokenCleartkExtractors.add(posExtraction1);
   
//    this.contextFeatureExtractors.add(new CleartkExtractor(IdentifiedAnnotation.class,
//        new CoveredTextExtractor(),
//        //new TypePathExtractor(IdentifiedAnnotation.class, "stem"),
//        new Preceding(2),
//        new Following(2)));
   
    // stab at dependency-based features
    //List<Feature> features = new ArrayList<Feature>();
    //ConllDependencyNode node1 = findAnnotationHead(jCas, arg1);

    CombinedExtractor baseExtractorCuePhraseCategory =
        new CombinedExtractor
          (
           new CoveredTextExtractor(),
           new TypePathExtractor(AssertionCuePhraseAnnotation.class, "cuePhrase"),
           new TypePathExtractor(AssertionCuePhraseAnnotation.class, "cuePhraseCategory"),
           new TypePathExtractor(AssertionCuePhraseAnnotation.class, "cuePhraseAssertionFamily")
          );
   
    cuePhraseInWindowExtractor = new CleartkExtractor(
        BaseToken.class,
        new CoveredTextExtractor(),
        new CleartkExtractor.Bag(new CleartkExtractor.Covered())
//          AssertionCuePhraseAnnotation.class,
//          baseExtractorCuePhraseCategory,
View Full Code Here
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
           new TypePathExtractor(AssertionCuePhraseAnnotation.class, "cuePhrase"),
           new TypePathExtractor(AssertionCuePhraseAnnotation.class, "cuePhraseCategory"),
           new TypePathExtractor(AssertionCuePhraseAnnotation.class, "cuePhraseAssertionFamily")
          );
   
    CleartkExtractor cuePhraseInWindowExtractor =
        new CleartkExtractor(
              AssertionCuePhraseAnnotation.class,
              baseExtractorCuePhraseCategory,
              new CleartkExtractor.Bag(new CleartkExtractor.Preceding(5)),
              new CleartkExtractor.Bag(new CleartkExtractor.Following(5))
              );
//              new CleartkExtractor.Ngram(new CleartkExtractor.Preceding(5), new CleartkExtractor.Following(5)));
              //new CoveredTextExtractor(),
//              new CleartkExtractor.Covered());
//              new CleartkExtractor.Preceding(5),
//              new CleartkExtractor.Following(5));
   
    CleartkExtractor tokenExtraction1 =
        new CleartkExtractor(
            BaseToken.class,
            new CoveredTextExtractor(),
            //new CleartkExtractor.Covered(),
            new CleartkExtractor.LastCovered(2),
            new CleartkExtractor.Preceding(5),
            new CleartkExtractor.Following(4),
            new CleartkExtractor.Bag(new CleartkExtractor.Preceding(10)),
            new CleartkExtractor.Bag(new CleartkExtractor.Following(10))
            );
   
    CleartkExtractor posExtraction1 =
        new CleartkExtractor(
            BaseToken.class,
            new TypePathExtractor(BaseToken.class, "partOfSpeech"),
            new CleartkExtractor.LastCovered(2),
            new CleartkExtractor.Preceding(3),
            new CleartkExtractor.Following(2)
View Full Code Here
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
130
131
132
133
134
135
136
137
138
139
140
           new TypePathExtractor(AssertionCuePhraseAnnotation.class, "cuePhrase"),
           new TypePathExtractor(AssertionCuePhraseAnnotation.class, "cuePhraseCategory"),
           new TypePathExtractor(AssertionCuePhraseAnnotation.class, "cuePhraseAssertionFamily")
          );
   
    CleartkExtractor cuePhraseInWindowExtractor =
        new CleartkExtractor(
              AssertionCuePhraseAnnotation.class,
              baseExtractorCuePhraseCategory,
              new CleartkExtractor.Bag(new CleartkExtractor.Preceding(5)),
              new CleartkExtractor.Bag(new CleartkExtractor.Following(5))
              );
//              new CleartkExtractor.Ngram(new CleartkExtractor.Preceding(5), new CleartkExtractor.Following(5)));
              //new CoveredTextExtractor(),
//              new CleartkExtractor.Covered());
//              new CleartkExtractor.Preceding(5),
//              new CleartkExtractor.Following(5));
   
    List<CleartkExtractor> extractorList = new ArrayList<CleartkExtractor>();
    extractorList.add(cuePhraseInWindowExtractor);
   
    //NamingExtractor cuePhraseInWindowNamingExtractor = new NamingExtractor("cuePhraseCategory__", cuePhraseInWindowExtractor);

    Collection<IdentifiedAnnotation> identifiedAnnotations = JCasUtil.select(jcas,  IdentifiedAnnotation.class);
    for (IdentifiedAnnotation current : identifiedAnnotations)
    {
      if (!(current instanceof EntityMention) && !(current instanceof EventMention)) continue;
     
      // otherwise current is an entity or event mention...
      logger.info(String.format("identified annotation (event or entity) [%d-%d] \"%s\" [%s]", current.getBegin(), current.getEnd(), current.getCoveredText(), current.getClass().getName()));
     
      Collection<Sentence> coveringSentences = entityToSentenceMap.get(current);
      if (coveringSentences == null || coveringSentences.isEmpty())
      {
        logger.info("no covering sentences found!!! continuing with next entity/event...");
        continue;
      }
      logger.info(String.format("covering sentence count: %d", coveringSentences.size()));
      Sentence firstCoveringSentence = coveringSentences.iterator().next();
     
      logger.info(String.format(
          "first covering sentence: [%d-%d] \"%s\" (%s)",
          firstCoveringSentence.getBegin(), firstCoveringSentence.getEnd(),
          firstCoveringSentence.getCoveredText(),
          firstCoveringSentence.getClass().getName()));
     
      List<Feature> cuePhraseFeatures =
          //cuePhraseInSentenceExtractor.extract(jcas, firstCoveringSentence);
          cuePhraseInWindowExtractor.extractWithin(jcas, current, firstCoveringSentence);
          //cuePhraseInWindowNamingExtractor.extract(jcas, current);
      if (cuePhraseFeatures != null && !cuePhraseFeatures.isEmpty())
      {
        String featureDebugString = (cuePhraseFeatures == null) ? "(no cue phrase features)" : cuePhraseFeatures.toString();
        logger.info("### cue phrase features: " + featureDebugString);
View Full Code Here
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
    // a list of feature extractors that require the token and the sentence
//    this.contextFeatureExtractors = new ArrayList<CleartkExtractor>();
   
    this.tokenCleartkExtractors = new ArrayList<CleartkExtractor>();

    CleartkExtractor tokenExtraction1 =
        new CleartkExtractor(
            BaseToken.class,
//            new FeatureFunctionExtractor(new CoveredTextExtractor(), new LowerCaseFeatureFunction()),
            new CoveredTextExtractor(),
            //new CleartkExtractor.Covered(),
            new CleartkExtractor.LastCovered(2),
            new CleartkExtractor.Preceding(5),
            new CleartkExtractor.Following(4),
            new CleartkExtractor.Bag(new CleartkExtractor.Preceding(3)),
            new CleartkExtractor.Bag(new CleartkExtractor.Following(3)),
            new CleartkExtractor.Bag(new CleartkExtractor.Preceding(5)),
            new CleartkExtractor.Bag(new CleartkExtractor.Following(5)),
            new CleartkExtractor.Bag(new CleartkExtractor.Preceding(10)),
            new CleartkExtractor.Bag(new CleartkExtractor.Following(10))
            );
   
//    CleartkExtractor posExtraction1 =
//        new CleartkExtractor(
//            BaseToken.class,
//            new TypePathExtractor(BaseToken.class, "partOfSpeech"),
//            new CleartkExtractor.LastCovered(2),
//            new CleartkExtractor.Preceding(3),
//            new CleartkExtractor.Following(2)
//            );

    this.tokenCleartkExtractors.add(tokenExtraction1);
    //this.tokenCleartkExtractors.add(posExtraction1);
   
//    this.contextFeatureExtractors.add(new CleartkExtractor(IdentifiedAnnotation.class,
//        new CoveredTextExtractor(),
//        //new TypePathExtractor(IdentifiedAnnotation.class, "stem"),
//        new Preceding(2),
//        new Following(2)));
   
    // stab at dependency-based features
    //List<Feature> features = new ArrayList<Feature>();
    //ConllDependencyNode node1 = findAnnotationHead(jCas, arg1);

    CombinedExtractor baseExtractorCuePhraseCategory =
        new CombinedExtractor
          (
           new CoveredTextExtractor(),
           new TypePathExtractor(AssertionCuePhraseAnnotation.class, "cuePhrase"),
           new TypePathExtractor(AssertionCuePhraseAnnotation.class, "cuePhraseCategory"),
           new TypePathExtractor(AssertionCuePhraseAnnotation.class, "cuePhraseAssertionFamily")
          );
   
    cuePhraseInWindowExtractor = new CleartkExtractor(
        BaseToken.class,
        new CoveredTextExtractor(),
        new CleartkExtractor.Bag(new CleartkExtractor.Covered())
//          AssertionCuePhraseAnnotation.class,
//          baseExtractorCuePhraseCategory,
View Full Code Here
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
130
131
132
133
134
135
136
137
138
139
140
           new TypePathExtractor(AssertionCuePhraseAnnotation.class, "cuePhrase"),
           new TypePathExtractor(AssertionCuePhraseAnnotation.class, "cuePhraseCategory"),
           new TypePathExtractor(AssertionCuePhraseAnnotation.class, "cuePhraseAssertionFamily")
          );
   
    CleartkExtractor cuePhraseInWindowExtractor =
        new CleartkExtractor(
              AssertionCuePhraseAnnotation.class,
              baseExtractorCuePhraseCategory,
              new CleartkExtractor.Bag(new CleartkExtractor.Preceding(5)),
              new CleartkExtractor.Bag(new CleartkExtractor.Following(5))
              );
//              new CleartkExtractor.Ngram(new CleartkExtractor.Preceding(5), new CleartkExtractor.Following(5)));
              //new CoveredTextExtractor(),
//              new CleartkExtractor.Covered());
//              new CleartkExtractor.Preceding(5),
//              new CleartkExtractor.Following(5));
   
    List<CleartkExtractor> extractorList = new ArrayList<CleartkExtractor>();
    extractorList.add(cuePhraseInWindowExtractor);
   
    //NamingExtractor cuePhraseInWindowNamingExtractor = new NamingExtractor("cuePhraseCategory__", cuePhraseInWindowExtractor);

    Collection<IdentifiedAnnotation> identifiedAnnotations = JCasUtil.select(jcas,  IdentifiedAnnotation.class);
    for (IdentifiedAnnotation current : identifiedAnnotations)
    {
      if (!(current instanceof EntityMention) && !(current instanceof EventMention)) continue;
     
      // otherwise current is an entity or event mention...
      logger.info(String.format("identified annotation (event or entity) [%d-%d] \"%s\" [%s]", current.getBegin(), current.getEnd(), current.getCoveredText(), current.getClass().getName()));
     
      Collection<Sentence> coveringSentences = entityToSentenceMap.get(current);
      if (coveringSentences == null || coveringSentences.isEmpty())
      {
        logger.info("no covering sentences found!!! continuing with next entity/event...");
        continue;
      }
      logger.info(String.format("covering sentence count: %d", coveringSentences.size()));
      Sentence firstCoveringSentence = coveringSentences.iterator().next();
     
      logger.info(String.format(
          "first covering sentence: [%d-%d] \"%s\" (%s)",
          firstCoveringSentence.getBegin(), firstCoveringSentence.getEnd(),
          firstCoveringSentence.getCoveredText(),
          firstCoveringSentence.getClass().getName()));
     
      List<Feature> cuePhraseFeatures =
          //cuePhraseInSentenceExtractor.extract(jcas, firstCoveringSentence);
          cuePhraseInWindowExtractor.extractWithin(jcas, current, firstCoveringSentence);
          //cuePhraseInWindowNamingExtractor.extract(jcas, current);
      if (cuePhraseFeatures != null && !cuePhraseFeatures.isEmpty())
      {
        String featureDebugString = (cuePhraseFeatures == null) ? "(no cue phrase features)" : cuePhraseFeatures.toString();
        logger.info("### cue phrase features: " + featureDebugString);
View Full Code Here
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
           new TypePathExtractor(AssertionCuePhraseAnnotation.class, "cuePhrase"),
           new TypePathExtractor(AssertionCuePhraseAnnotation.class, "cuePhraseCategory"),
           new TypePathExtractor(AssertionCuePhraseAnnotation.class, "cuePhraseAssertionFamily")
          );
   
    CleartkExtractor cuePhraseInWindowExtractor =
        new CleartkExtractor(
              AssertionCuePhraseAnnotation.class,
              baseExtractorCuePhraseCategory,
              new CleartkExtractor.Bag(new CleartkExtractor.Preceding(5)),
              new CleartkExtractor.Bag(new CleartkExtractor.Following(5))
              );
//              new CleartkExtractor.Ngram(new CleartkExtractor.Preceding(5), new CleartkExtractor.Following(5)));
              //new CoveredTextExtractor(),
//              new CleartkExtractor.Covered());
//              new CleartkExtractor.Preceding(5),
//              new CleartkExtractor.Following(5));
   
    CleartkExtractor tokenExtraction1 =
        new CleartkExtractor(
            BaseToken.class,
            new CoveredTextExtractor(),
            //new CleartkExtractor.Covered(),
            new CleartkExtractor.LastCovered(2),
            new CleartkExtractor.Preceding(5),
            new CleartkExtractor.Following(4),
            new CleartkExtractor.Bag(new CleartkExtractor.Preceding(10)),
            new CleartkExtractor.Bag(new CleartkExtractor.Following(10))
            );
   
    CleartkExtractor posExtraction1 =
        new CleartkExtractor(
            BaseToken.class,
            new TypePathExtractor(BaseToken.class, "partOfSpeech"),
            new CleartkExtractor.LastCovered(2),
            new CleartkExtractor.Preceding(3),
            new CleartkExtractor.Following(2)
View Full Code Here
TOP

Related Classes of org.cleartk.classifier.feature.extractor.CleartkExtractor

  • org.apache.ctakes.assertion.medfacts.cleartk.AssertionCleartkAnalysisEngine
  • org.apache.ctakes.assertion.medfacts.cleartk.extractors.CueWordTest
  • org.apache.ctakes.assertion.medfacts.cleartk.extractors.CueWordTest2
  • org.apache.ctakes.temporal.ae.BackwardsTimeAnnotator
  • org.apache.ctakes.temporal.ae.ConstituencyBasedTimeAnnotator
  • org.apache.ctakes.temporal.ae.ContextualModalityAnnotator
  • org.apache.ctakes.temporal.ae.CRFTimeAnnotator
  • org.apache.ctakes.temporal.ae.DocTimeRelAnnotator
  • org.apache.ctakes.temporal.ae.EventAdmissionTimeAnnotator
  • org.apache.ctakes.temporal.ae.EventAnnotator
    • 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.