edu.cmu.minorthird.classify.ranking
Class RankingBoosted

java.lang.Object
  extended by edu.cmu.minorthird.classify.BatchClassifierLearner
      extended by edu.cmu.minorthird.classify.BatchBinaryClassifierLearner
          extended by edu.cmu.minorthird.classify.ranking.BatchRankingLearner
              extended by edu.cmu.minorthird.classify.ranking.RankingBoosted
All Implemented Interfaces:
BinaryClassifierLearner, ClassifierLearner, java.lang.Cloneable

public class RankingBoosted
extends BatchRankingLearner

A boosted version for ranking. An implementation of - "Michael Collins and Terry Koo, Discriminative Reranking for Natural Language Parsing. Computational Linguistics, March 2005", see page 45. Requirements of this class: - Considers only binary features - Requires a particular cont. feature named "walkerScore", that contains the original log-probability for an example To do: - Handle multiple positive answers. - Automatically discretize real-value features into binary. - It is possible to incorporate example 'importance' weights, according to some 'goodness' evaluation measure. See Collins'.

Author:
Einat Minkov

Field Summary
 
Fields inherited from class edu.cmu.minorthird.classify.BatchClassifierLearner
classifier, dataset
 
Constructor Summary
RankingBoosted()
           
RankingBoosted(int numEpochs, int exampleSize)
           
 
Method Summary
 Classifier batchTrain(Dataset data)
          subclasses should use this method to implement a batch supervised learning algorithm.
 double initialExpLoss(double w0, Example[][] rankedExamples)
           
 
Methods inherited from class edu.cmu.minorthird.classify.ranking.BatchRankingLearner
listsWithOneExampleEach, sortByScore, splitIntoRankings
 
Methods inherited from class edu.cmu.minorthird.classify.BatchBinaryClassifierLearner
batchTrainBinary, getBinaryClassifier, getSchema, setSchema
 
Methods inherited from class edu.cmu.minorthird.classify.BatchClassifierLearner
addExample, completeTraining, copy, getClassifier, hasNextQuery, nextQuery, reset, setInstancePool
 
Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait
 
Methods inherited from interface edu.cmu.minorthird.classify.ClassifierLearner
addExample, completeTraining, copy, getClassifier, hasNextQuery, nextQuery, reset, setInstancePool
 

Constructor Detail

RankingBoosted

public RankingBoosted()

RankingBoosted

public RankingBoosted(int numEpochs,
                      int exampleSize)
Method Detail

batchTrain

public Classifier batchTrain(Dataset data)
Description copied from class: BatchClassifierLearner
subclasses should use this method to implement a batch supervised learning algorithm.

Specified by:
batchTrain in class BatchClassifierLearner

initialExpLoss

public double initialExpLoss(double w0,
                             Example[][] rankedExamples)