Package org.mitre.medfacts.i2b2.processors

Examples of org.mitre.medfacts.i2b2.processors.ScopeFileProcessor

965
966
967
968
969
970
971
972
973
974
975
      AssertionAnnotation assertForTI = trainingInstance.getAssertAnnotateForTI();
      //Count number of enclosing negation and speculation scopes
      for (ScopeAnnotation enclosingScope : assertForTI.getEnclosingScopes())
      {
        CueAnnotation cueForScope = enclosingScope.getCueForScope();
        CueSubType scopeType = cueForScope.getCueSubType();
        if (scopeType == CueSubType.NEGATION) enclosingNegationScopeCnt++;
        else if (scopeType == CueSubType.SPECULATION) enclosingSpeculationScopeCnt++;
        else logger.info(String.format("WARNING: CUE %s%n  FOR SCOPE %s%n ENCLOSING %s%n is neither a negation nor speculation cue%n", cueForScope, enclosingScope,assertForTI));
      }
      if (checkForEnabledFeature("statusRuleMixNMatchFeature"))
View Full Code Here
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
        if (annotationsAtCurrentPosition != null)
        for (Annotation a : annotationsAtCurrentPosition)
        {
          if (a instanceof CueWordAnnotation)
          {
            CueWordAnnotation cueWord = (CueWordAnnotation)a;
            String cueWordType = cueWord.getCueWordType().toString();
            if (checkForEnabledFeature("cueWord"))
            {
              int cueWordBegin = cueWord.getBegin().getTokenOffset();
              int cueWordEnd = cueWord.getEnd().getTokenOffset();
              if (cueWordBegin < conceptBeginTokenOffset) {
                  trainingInstance.addFeature("cueWord_" + cueWordType + "_left");
                  if ((conceptBeginTokenOffset - cueWordBegin) < 4) {
                      trainingInstance.addFeature("cueWord_" + cueWordType + "_left_3");
                  }
              } else if (cueWordBegin > conceptEndTokenOffset) {
                    trainingInstance.addFeature("cueWord_" + cueWordType + "_right");
                    if ((cueWordEnd - conceptEndTokenOffset) < 4) {
                        trainingInstance.addFeature("cueWord_" + cueWordType + "_right_3");
                    }
              } else {
                  trainingInstance.addFeature("cueWord_" + cueWordType + "_within");
              }
            }
            if (checkForEnabledFeature("cueWordValue"))
            {
              trainingInstance.addFeature("cueword_" + cueWord.getCueWordText());
            }
          }
          if (checkForEnabledFeature("concepts")) {
            if (a instanceof ConceptAnnotation) {
                ConceptAnnotation concept = (ConceptAnnotation) a;
View Full Code Here
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
      if (checkForEnabledFeature("cueWordOrderingsLeft")) {
          List<CueWordAnnotation> annots = new ArrayList<CueWordAnnotation>();
          for (Annotation a : allLineAnnotations) {
      if (a instanceof CueWordAnnotation) {
        CueWordAnnotation an = (CueWordAnnotation)a;
        if ((inCWSCueWordSet(an.getCueWordType())) && (an.getBegin().getTokenOffset() < conceptBeginTokenOffset)) {
                  annots.add(an);
              }
      }
          }
    if (annots.size() > 0) {
      Collections.sort(annots);
      StringBuilder str = new StringBuilder("CWS_left");
      for (CueWordAnnotation a : annots) {
        str.append("_");
        str.append(a.getCueWordType());
      }
      trainingInstance.addFeature(str.toString());
    }
      }

      if (checkForEnabledFeature("cueWordOrderingsRight")) {
          List<CueWordAnnotation> annots = new ArrayList<CueWordAnnotation>();
          for (Annotation a : allLineAnnotations) {
      if (a instanceof CueWordAnnotation) {
        CueWordAnnotation an = (CueWordAnnotation)a;
        if ((inCWSCueWordSet(an.getCueWordType())) && (an.getBegin().getTokenOffset() > conceptEndTokenOffset)) {
    annots.add(an);
              }
      }
          }
          Collections.sort(annots);
    if (annots.size() > 0) {
      StringBuilder str = new StringBuilder("CWS_right");
      for (CueWordAnnotation a : annots) {
              str.append("_");
              str.append(a.getCueWordType());
      }
      trainingInstance.addFeature(str.toString());
    }
      }

      //logger.info(String.format("lineNumber: %d%n", lineNumber);
      String tokensOnCurrentLine[] = textLookup[lineNumber-1];
      for (int currentTokenOffset=0; currentTokenOffset < tokensOnCurrentLine.length; currentTokenOffset++)
      {
        String currentToken = tokensOnCurrentLine[currentTokenOffset];
        List<Annotation> annotationsAtCurrentPosition = indexer.findAnnotationsForPosition(lineNumber, currentTokenOffset);

        int scopeCount = 0;
        if (annotationsAtCurrentPosition != null)
        for (Annotation a : annotationsAtCurrentPosition)
        {
            if (checkForEnabledFeature("concepts")) {
            if (a instanceof ConceptAnnotation) {
                ConceptAnnotation concept = (ConceptAnnotation) a;

                String conceptType = concept.getConceptType().toString();
                int thisConceptBegin = concept.getBegin().getTokenOffset();
                int thisConceptEnd = concept.getEnd().getTokenOffset();
                if (concept.getBegin().getTokenOffset() < conceptBeginTokenOffset) {
                    trainingInstance.addFeature("concept_" + conceptType + "_left");
                    if ((conceptBeginTokenOffset - thisConceptEnd) < 4) {
                        trainingInstance.addFeature("concept_" + conceptType + "_left_3");
                    }
                } else {
                    if ((thisConceptBegin - conceptEndTokenOffset) < 4) {
                        trainingInstance.addFeature("concept_" + conceptType + "_right_3");
                    }
                    trainingInstance.addFeature("concept_" + conceptType + "_right");
                }
            }
            }

          if (a instanceof ScopeAnnotation)
          {
            ScopeAnnotation scope = (ScopeAnnotation)a;
            scopeCount++;
            if (checkForEnabledFeature("scope"))
            {
              trainingInstance.addFeature("scope");
            }
            if (checkForEnabledFeature("inScope"))
            {
              trainingInstance.addFeature("in_scope_" + currentToken);
            }
            if (checkForEnabledFeature("inScopeId"))
            {
              trainingInstance.addFeature("in_scope_id_" + scope.getScopeId() + "_" + currentToken);
            }
          }

          if (a instanceof CueAnnotation)
          {
            CueAnnotation cue = (CueAnnotation)a;
            if (checkForEnabledFeature("cue"))
            {
              String cueType = cue.getCueSubType().toString();
              int cueBegin = cue.getBegin().getTokenOffset();
              if (cueBegin < conceptBeginTokenOffset) {
                  trainingInstance.addFeature("cue_" + cueType + "_left");
                  if ((conceptBeginTokenOffset - cueBegin) < 4) {
                      trainingInstance.addFeature("cue_" + cueType + "_left_3");
                  }
              } else {
                  int cueEnd = cue.getEnd().getTokenOffset();
                  trainingInstance.addFeature("cue_" + cueType + "_right");
                  if ((cueEnd - conceptEndTokenOffset) < 4) {
                      trainingInstance.addFeature("cue_" + cueType + "_right_3");
                  }
              }
            }
            if (checkForEnabledFeature("inCue"))
            {
              trainingInstance.addFeature("in_cue_" + currentToken);
            }
            if (checkForEnabledFeature("inCueForScopeId"))
            {
              trainingInstance.addFeature("in_cue_for_scope_id_" + cue.getScopeIdReference() + "_" + currentToken);
            }
          }

          if (a instanceof CueWordAnnotation)
          {
            CueWordAnnotation cueWord = (CueWordAnnotation)a;
            String cueWordType = cueWord.getCueWordType().toString();
            String cueWordText = cueWord.getCueWordText();
            String escapedCueWordText = escapeFeatureName(cueWordText);
            String escapedCueWordClass = null;
            String cueWordClass = cueWord.getCueWordClass();
            boolean cueWordClassIsNotEmpty = (cueWordClass != null) && (!cueWordClass.isEmpty());
            if (checkForEnabledFeature("cueWordClassValue") && cueWordClassIsNotEmpty)
            {
              escapedCueWordClass = escapeFeatureName(cueWordClass);
            }

            if (checkForEnabledFeature("cueWordTextPositional"))
            {
              int cueWordBegin = cueWord.getBegin().getTokenOffset();
              int cueWordEnd = cueWord.getEnd().getTokenOffset();
              if (cueWordBegin < conceptBeginTokenOffset) {
                  trainingInstance.addFeature("cueWordTextPositional_" + escapedCueWordText + "_left");
                  if ((conceptBeginTokenOffset - cueWordBegin) < 4) {
                      trainingInstance.addFeature("cueWordTextPositional_" + escapedCueWordText + "_left_3");
                  }
              } else if (cueWordBegin > conceptEndTokenOffset) {
                    trainingInstance.addFeature("cueWordTextPositional_" + escapedCueWordText + "_right");
                    if ((cueWordEnd - conceptEndTokenOffset) < 4) {
                        trainingInstance.addFeature("cueWordTextPositional_" + escapedCueWordText + "_right_3");
                    }
              } else {
                  trainingInstance.addFeature("cueWordTextPositional_" + escapedCueWordText + "_within");
              }
            }
            if (checkForEnabledFeature("cueWordTypePositional"))
            {
              int cueWordBegin = cueWord.getBegin().getTokenOffset();
              int cueWordEnd = cueWord.getEnd().getTokenOffset();
              if (cueWordBegin < conceptBeginTokenOffset) {
                  trainingInstance.addFeature("cueWordTypePositional_" + cueWordType + "_left");
                  if ((conceptBeginTokenOffset - cueWordBegin) < 4) {
                      trainingInstance.addFeature("cueWordTypePositional_" + cueWordType + "_left_3");
                  }
              } else if (cueWordBegin > conceptEndTokenOffset) {
                    trainingInstance.addFeature("cueWordTypePositional_" + cueWordType + "_right");
                    if ((cueWordEnd - conceptEndTokenOffset) < 4) {
                        trainingInstance.addFeature("cueWordTypePositional_" + cueWordType + "_right_3");
                    }
              } else {
                  trainingInstance.addFeature("cueWordTypePositional_" + cueWordType + "_within");
              }
            }
            if (checkForEnabledFeature("cueWordClassPositional") && cueWordClassIsNotEmpty)
            {
              int cueWordBegin = cueWord.getBegin().getTokenOffset();
              int cueWordEnd = cueWord.getEnd().getTokenOffset();
              if (cueWordBegin < conceptBeginTokenOffset) {
                  trainingInstance.addFeature("cueWordClassPositional_" + escapedCueWordClass + "_left");
                  if ((conceptBeginTokenOffset - cueWordBegin) < 4) {
                      trainingInstance.addFeature("cueWordClassPositional_" + escapedCueWordClass + "_left_3");
                  }
              } else if (cueWordBegin > conceptEndTokenOffset) {
                    trainingInstance.addFeature("cueWordClassPositional_" + escapedCueWordClass + "_right");
                    if ((cueWordEnd - conceptEndTokenOffset) < 4) {
                        trainingInstance.addFeature("cueWordClassPositional_" + escapedCueWordClass + "_right_3");
                    }
              } else {
                  trainingInstance.addFeature("cueWordClassPositional_" + escapedCueWordClass + "_within");
              }
            }
            if (checkForEnabledFeature("cueWordTextValue"))
            {
              trainingInstance.addFeature("cueWordTextValue_" + escapedCueWordText);
            }
            if (checkForEnabledFeature("cueWordTypeValue") && cueWord.getCueWordType() != null)
            {
              trainingInstance.addFeature("cueWordTypeValue_" + cueWord.getCueWordType().toString());
            }
            if (checkForEnabledFeature("cueWordClassValue") && cueWord.getCueWordClass() != null && !cueWord.getCueWordClass().isEmpty())
            {
              trainingInstance.addFeature("cueWordClassValue_" + escapedCueWordClass);
            }
          }
View Full Code Here
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
        {
          boolean matched = compareForCueMatch(cueItem, currentLine, tokenOffset);
          if (matched)
          {
            logger.finest(String.format("MATCHES!!!%n  BEGIN MATCH%n  cue: %s%n  inputLine: %s%n  position: %d%n  END MATCH%n%n", cueItem.toString(), convertLineToString(currentLine), tokenOffset));
            CueWordAnnotation a = new CueWordAnnotation();
            int beginOffset = tokenOffset;
            int endOffset = tokenOffset + size - 1;
            Location beginLocation = new Location();
            beginLocation.setLine(lineNumberOffset + 1);
            beginLocation.setTokenOffset(beginOffset);
            Location endLocation = new Location();
            endLocation.setLine(lineNumberOffset + 1);
            endLocation.setTokenOffset(endOffset);
            a.setBegin(beginLocation);
            a.setEnd(endLocation);
            a.setCueWordType(getCueWordType());
            a.setCueWordClass(cueItem.getCueWordClass());
            a.setCueWordText(constructWordSequenceText(currentLine, beginOffset, endOffset));
            annotationList.add(a);
          }
        }

        // generate alternative sentences for the current line with any
View Full Code Here
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
    try
    {
      for (Object[] current : temp)
      {
        CueWordType cueWordType = (CueWordType)current[0];
        String cueFilename = (String)current[1];
        MedFactsRunner.processCueList(cueFilename, cueWordType, textLookup, allAnnotationList, annotationsByType);
      }
    }
    catch (URISyntaxException e)
View Full Code Here
32
33
34
35
36
37
38
39
40
41
42
43
    {
      List<Annotation> annotationsAtCurrentPosition = indexer.findAnnotationsForPosition(lineNumber, i);
      if (annotationsAtCurrentPosition != null) {
        for (Annotation a : annotationsAtCurrentPosition) {
    if (a instanceof PartOfSpeechAnnotation) {
      PartOfSpeechAnnotation posAn = (PartOfSpeechAnnotation)a;
      sb.append(posAn.getPartOfSpeech());
          }
  }
      }
    }
    return StringHandling.escapeStringForFeatureName(sb.toString());
View Full Code Here
53
54
55
56
57
58
59
60
61
62
63
64
    {
      List<Annotation> annotationsAtCurrentPosition = indexer.findAnnotationsForPosition(lineNumber, i);
      if (annotationsAtCurrentPosition != null) {
        for (Annotation a : annotationsAtCurrentPosition) {
    if (a instanceof PartOfSpeechAnnotation) {
      PartOfSpeechAnnotation posAn = (PartOfSpeechAnnotation)a;
      sb.append(posAn.getPartOfSpeech());
          }
  }
      }
    }
    return StringHandling.escapeStringForFeatureName(sb.toString());
View Full Code Here
121
122
123
124
125
126
127
128
129
130
131
      ScopeParser scopeParser = new ScopeParser(scopeModelFilePath,
          cueModelFilePath);
      assertionDecoderConfiguration.setScopeParser(scopeParser);

      logger.info(String.format("pos model file: %s", posModelFilePath));
      PartOfSpeechTagger posTagger = new PartOfSpeechTagger(posModelFilePath);
      assertionDecoderConfiguration.setPosTagger(posTagger);

      Set<String> enabledFeatureIdSet = null;
      enabledFeatureIdSet = BatchRunner
          .loadEnabledFeaturesFromFile(enabledFeaturesFile);
View Full Code Here
120
121
122
123
124
125
126
127
128
129
130
      ScopeParser scopeParser = new ScopeParser(scopeModelFilePath,
          cueModelFilePath);
      assertionDecoderConfiguration.setScopeParser(scopeParser);

      logger.info(String.format("pos model file: %s", posModelFilePath));
      PartOfSpeechTagger posTagger = new PartOfSpeechTagger(posModelFilePath);
      assertionDecoderConfiguration.setPosTagger(posTagger);

      Set<String> enabledFeatureIdSet = null;
      enabledFeatureIdSet = BatchRunner
          .loadEnabledFeaturesFromFile(enabledFeaturesFile);
View Full Code Here
230
231
232
233
234
235
236
237
238
239
240
    }

    logger.info(String.format("cue model file: %s%n", cueModelFile.getAbsolutePath()));
    //initialize scope/cue parser
    ScopeParser scopeParser = new ScopeParser(scopeModelFile.getAbsolutePath(), cueModelFile.getAbsolutePath());
    PartOfSpeechTagger posTagger = new PartOfSpeechTagger(posModelFile.getAbsolutePath());

//    String baseDirectory = args[0];
//    logger.info(String.format("base directory: %s%n", baseDirectory);

    BatchRunner batchRunner = new BatchRunner();
View Full Code Here
TOP

Related Classes of org.mitre.medfacts.i2b2.processors.ScopeFileProcessor

  • edu.mayo.bmi.uima.core.type.syntax.BaseToken
  • edu.mayo.bmi.uima.core.type.textspan.Sentence
  • org.apache.commons.cli.CommandLine
  • org.apache.commons.cli.CommandLineParser
  • org.apache.commons.cli.GnuParser
  • org.apache.commons.cli.Options
  • org.apache.ctakes.assertion.medfacts.AssertionAnalysisEngine
  • org.apache.ctakes.assertion.medfacts.ConceptConverterAnalysisEngine
  • org.apache.ctakes.assertion.medfacts.ConceptLookup
  • org.apache.ctakes.assertion.medfacts.Converti2b2AnnotationsToCTAKES
    • 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.