edu.illinois.cs.cogcomp.lbj.coref.decoders
Class BIODecoder

java.lang.Object
  extended by edu.illinois.cs.cogcomp.lbj.coref.decoders.DecoderWithOptions<ChunkSolution>
      extended by edu.illinois.cs.cogcomp.lbj.coref.decoders.ChunkDecoder
          extended by edu.illinois.cs.cogcomp.lbj.coref.decoders.BIODecoder
All Implemented Interfaces:
SolutionDecoder<ChunkSolution>

public class BIODecoder
extends ChunkDecoder

Translates a BIO classifier's decisions into a ChunkSolution.


Field Summary
 
Fields inherited from class edu.illinois.cs.cogcomp.lbj.coref.decoders.ChunkDecoder
m_classifier
 
Fields inherited from class edu.illinois.cs.cogcomp.lbj.coref.decoders.DecoderWithOptions
m_options, m_train
 
Constructor Summary
BIODecoder(LBJ2.classify.Classifier bioClassifier)
          Constructs the decoder.
 
Method Summary
 ChunkSolution decode(Doc doc)
          Classifies each word (wrapped in a BIOExample) in a document, starting at the first text word, as beginning ('b'), inside ('i') or outside ('o') a chunk, and decodes these decisions into a list of chunks.
 
Methods inherited from class edu.illinois.cs.cogcomp.lbj.coref.decoders.DecoderWithOptions
getBooleanOption, getOption, getRealOption, processOption, setOption, setOption, setOption, setTrain
 
Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait
 

Constructor Detail

BIODecoder

public BIODecoder(LBJ2.classify.Classifier bioClassifier)
Constructs the decoder.

Parameters:
bioClassifier - A classifier that classifies a word (wrapped in a BIOExample), as beginning ("b"), inside ("i") or outside ("o") a chunk.
Method Detail

decode

public ChunkSolution decode(Doc doc)
Classifies each word (wrapped in a BIOExample) in a document, starting at the first text word, as beginning ('b'), inside ('i') or outside ('o') a chunk, and decodes these decisions into a list of chunks.

Specified by:
decode in interface SolutionDecoder<ChunkSolution>
Specified by:
decode in class ChunkDecoder
Parameters:
doc - The document to process.
Returns:
A ChunkSolution containing a list of the detected chunks.