cc.mallet.fst.confidence
Class ConfidenceCorrectorEvaluator

java.lang.Object
  extended by cc.mallet.fst.confidence.ConfidenceCorrectorEvaluator

public class ConfidenceCorrectorEvaluator
extends java.lang.Object

Calculates the effectiveness of "constrained viterbi" in propagating corrections in one segment of a sequence to other segments.


Constructor Summary
ConfidenceCorrectorEvaluator(java.lang.Object[] startTags, java.lang.Object[] inTags)
           
 
Method Summary
 void evaluate(Transducer model, java.util.ArrayList predictions, InstanceList ilist, java.util.ArrayList correctedSegments, java.lang.String description, java.io.PrintStream outputStream, boolean errorsInUncorrected)
          Only evaluates over sequences which contain errors.
 
Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait
 

Constructor Detail

ConfidenceCorrectorEvaluator

public ConfidenceCorrectorEvaluator(java.lang.Object[] startTags,
                                    java.lang.Object[] inTags)
Method Detail

evaluate

public void evaluate(Transducer model,
                     java.util.ArrayList predictions,
                     InstanceList ilist,
                     java.util.ArrayList correctedSegments,
                     java.lang.String description,
                     java.io.PrintStream outputStream,
                     boolean errorsInUncorrected)
Only evaluates over sequences which contain errors. Examine region not directly corrected by correctedSegments to measure effects of error propagation.

Parameters:
model - used to segment input sequence
predictions - list of the corrected segmentation
ilist - list of testing data
correctedSegments - list of Segments in each sequence that were corrected...currently only allows one segment per instance.
uncorrected - true if we only evaluate sequences where errors remain after correction