org.cleartk.token.tokenizer.chunk
Class ChunkTokenizerLabeler

java.lang.Object
  org.cleartk.chunker.ChunkLabeler_ImplBase
      org.cleartk.chunker.DefaultChunkLabeler
          org.cleartk.token.tokenizer.chunk.ChunkTokenizerLabeler

All Implemented Interfaces:

org.cleartk.chunker.ChunkLabeler, org.uimafit.factory.initializable.Initializable

Deprecated.

@Deprecated
public class ChunkTokenizerLabeler
extends org.cleartk.chunker.DefaultChunkLabeler

Copyright (c) 2007-2008, Regents of the University of Colorado
All rights reserved.

Field Summary

Fields inherited from class org.cleartk.chunker.DefaultChunkLabeler

PARAM_CHUNK_LABEL_FEATURE_NAME

Fields inherited from class org.cleartk.chunker.ChunkLabeler_ImplBase

annotationLabels, BEGIN_PREFIX, chunkAnnotationClass, chunkAnnotationType, INSIDE_PREFIX, labeledAnnotationClass, labeledAnnotationType, OUTSIDE_LABEL, PARAM_CHUNK_ANNOTATION_CLASS_NAME, SEPARATOR, typesInitialized

Constructor Summary

ChunkTokenizerLabeler()
Deprecated.

Method Summary

void	chunks2Labels(org.apache.uima.jcas.JCas jCas) Deprecated. In general, the chunks will be annotations of type Token and the labeled annotations will be annotations of type Subtoken creating labels such as B-TOKEN or I-TOKEN.

Methods inherited from class org.cleartk.chunker.DefaultChunkLabeler

createChunk, getChunkLabel, initialize, initializeTypes, setChunkLabelFeatureName

Methods inherited from class org.cleartk.chunker.ChunkLabeler_ImplBase

chunks2Labels, getLabel, labels2Chunks, setChunkAnnotationClassName, setLabel, setLabeledAnnotationClassName

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

ChunkTokenizerLabeler

public ChunkTokenizerLabeler()

Deprecated.

Method Detail

chunks2Labels

public void chunks2Labels(org.apache.uima.jcas.JCas jCas)
                   throws org.apache.uima.analysis_engine.AnalysisEngineProcessException

Deprecated.

In general, the chunks will be annotations of type Token and the labeled annotations will be annotations of type Subtoken creating labels such as B-TOKEN or I-TOKEN. I had to override chunk2Labels for the chunker-style tokenization because there are cases in the GENIA data where the "Token" has no Subtokens within it because they split a word like "heterodimer" into two tokens and the Subtokenizer only identifies one Subtoken from that word. So, when the Token corresponds to "hetero" it will label the Subtoken that contains it as "B-TOKEN". When the next Token corresponding to "dimer" is visited the same Subtoken that contains it will be relabeled as "B-TOKEN".

Throws:

org.apache.uima.analysis_engine.AnalysisEngineProcessException