edu.cmu.minorthird.classify.transform
Class LeaveOneOutDictTransformLearner

java.lang.Object
  edu.cmu.minorthird.classify.transform.LeaveOneOutDictTransformLearner

public class LeaveOneOutDictTransformLearner
extends java.lang.Object

Construct a transformation of a dataset that includes "leave one out dictionary".

The value of some feature for of all training examples will be stored in a soft dictionary, and distance to the closest dictionary entry will be used as an additional feature. For more information, see Sarawagi and Cohen, "Semi-Markov Conditional Random Fields for Information Extraction", 2004.

Author:

William Cohen

Field Summary

static java.lang.String[]

DEFAULT_PATTERN

Constructor Summary

LeaveOneOutDictTransformLearner()

LeaveOneOutDictTransformLearner(java.lang.String distanceNames)

LeaveOneOutDictTransformLearner(java.lang.String[] featurePattern)

LeaveOneOutDictTransformLearner(java.lang.String[] featurePattern, java.lang.String distanceNames)

Method Summary

InstanceTransform	batchTrain(Dataset dataset) Examine data, build an instance transformer
void	setSchema(ExampleSchema schema)
void	trainDistances(ExampleSchema schema, com.wcohen.ss.lookup.SoftDictionary[] softDict)

Methods inherited from class java.lang.Object

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

Field Detail

DEFAULT_PATTERN

public static final java.lang.String[] DEFAULT_PATTERN

Constructor Detail

LeaveOneOutDictTransformLearner

public LeaveOneOutDictTransformLearner()

LeaveOneOutDictTransformLearner

public LeaveOneOutDictTransformLearner(java.lang.String distanceNames)

LeaveOneOutDictTransformLearner

public LeaveOneOutDictTransformLearner(java.lang.String[] featurePattern)

LeaveOneOutDictTransformLearner

public LeaveOneOutDictTransformLearner(java.lang.String[] featurePattern,
                                       java.lang.String distanceNames)

Method Detail

setSchema

public void setSchema(ExampleSchema schema)

trainDistances

public void trainDistances(ExampleSchema schema,
                           com.wcohen.ss.lookup.SoftDictionary[] softDict)

batchTrain

public InstanceTransform batchTrain(Dataset dataset)

Examine data, build an instance transformer