cc.mallet.fst.confidence
Class MaxEntSequenceConfidenceEstimator

java.lang.Object
  extended by cc.mallet.fst.confidence.TransducerSequenceConfidenceEstimator
      extended by cc.mallet.fst.confidence.MaxEntSequenceConfidenceEstimator

public class MaxEntSequenceConfidenceEstimator
extends TransducerSequenceConfidenceEstimator

Estimates the confidence of a Sequence extracted by a Transducer using a MaxEnt classifier to classify Sequences as "correct" or "incorrect." xxx needs some interface work.


Field Summary
 
Fields inherited from class cc.mallet.fst.confidence.TransducerSequenceConfidenceEstimator
model
 
Constructor Summary
MaxEntSequenceConfidenceEstimator(Transducer model)
           
MaxEntSequenceConfidenceEstimator(Transducer model, double gaussianVariance)
           
 
Method Summary
 double estimateConfidenceFor(Instance instance, java.lang.Object[] startTags, java.lang.Object[] inTags)
          Calculates the confidence in the tagging of an Instance.
 MaxEnt getClassifier()
           
 PipedInstanceWithConfidence[] rankPipedInstancesByConfidence(InstanceList ilist, java.lang.Object[] startTags, java.lang.Object[] continueTags)
           
 MaxEnt trainClassifier(InstanceList ilist, java.lang.String correct, java.lang.String incorrect)
          Train underlying classifier on ilist.
 
Methods inherited from class cc.mallet.fst.confidence.TransducerSequenceConfidenceEstimator
rankInstancesByConfidence
 
Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait
 

Constructor Detail

MaxEntSequenceConfidenceEstimator

public MaxEntSequenceConfidenceEstimator(Transducer model,
                                         double gaussianVariance)

MaxEntSequenceConfidenceEstimator

public MaxEntSequenceConfidenceEstimator(Transducer model)
Method Detail

getClassifier

public MaxEnt getClassifier()

trainClassifier

public MaxEnt trainClassifier(InstanceList ilist,
                              java.lang.String correct,
                              java.lang.String incorrect)
Train underlying classifier on ilist. Assumes ilist has targst correct or incorrect.

Parameters:
ilist - training list to build correct/incorrect classifier
correct - "correct" label
incorrect - "incorrect" label

estimateConfidenceFor

public double estimateConfidenceFor(Instance instance,
                                    java.lang.Object[] startTags,
                                    java.lang.Object[] inTags)
Calculates the confidence in the tagging of an Instance.

Specified by:
estimateConfidenceFor in class TransducerSequenceConfidenceEstimator

rankPipedInstancesByConfidence

public PipedInstanceWithConfidence[] rankPipedInstancesByConfidence(InstanceList ilist,
                                                                    java.lang.Object[] startTags,
                                                                    java.lang.Object[] continueTags)