dragon.ml.seqmodel.feature

Class FeatureTypeWord

java.lang.Object

dragon.ml.seqmodel.feature.AbstractFeatureType

dragon.ml.seqmodel.feature.FeatureTypeWord

All Implemented Interfaces:

FeatureType

public class FeatureTypeWord
extends AbstractFeatureType

World feature type

Copyright: Copyright (c) 2005

Company: IST, Drexel University

Version:

1.0 If feature dictionary is given, no training is needed and all states are simply emitted for each word. Otherwise training is needed.

Author:

Davis Zhou

Field Summary

Fields

Modifier and Type	Field and Description
static int	RARE_THRESHOLD

Fields inherited from class dragon.ml.seqmodel.feature.AbstractFeatureType

idPrefix

Constructor Summary

Constructors

Constructor and Description
FeatureTypeWord(FeatureDictionary dict, int stateNum)
FeatureTypeWord(FeatureDictionary dict, int stateNum, boolean caseSensitive)
FeatureTypeWord(String wordFile, int stateNum)
FeatureTypeWord(String wordFile, int stateNum, boolean caseSensitive)

Method Summary

Modifier and Type	Method and Description
FeatureDictionary	getWordDictionary()
boolean	hasNext() Tests if next feature exists
Feature	next() Gets the next feature
boolean	readTrainingResult() Reads the training result in testing mode
boolean	saveTrainingResult() Saves the training results in training mode
boolean	startScanFeaturesAt(DataSequence data, int startPos, int endPos) Starts scanning features of the given segment of a sequence
boolean	supportSegment() Tests if the feature type supports generating features for segments.
boolean	train(Dataset trainData) Trains the feature types with the labeled training dataset

Methods inherited from class dragon.ml.seqmodel.feature.AbstractFeatureType

getTypeID, needTraining, setTypeID, startScanFeaturesAt

Methods inherited from class java.lang.Object

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

Field Detail

RARE_THRESHOLD

public static int RARE_THRESHOLD

Constructor Detail

FeatureTypeWord

public FeatureTypeWord(String wordFile,
                       int stateNum)

FeatureTypeWord

public FeatureTypeWord(String wordFile,
                       int stateNum,
                       boolean caseSensitive)

FeatureTypeWord

public FeatureTypeWord(FeatureDictionary dict,
                       int stateNum)

FeatureTypeWord

public FeatureTypeWord(FeatureDictionary dict,
                       int stateNum,
                       boolean caseSensitive)

Method Detail

getWordDictionary

public FeatureDictionary getWordDictionary()

startScanFeaturesAt

public boolean startScanFeaturesAt(DataSequence data,
                                   int startPos,
                                   int endPos)

Description copied from interface: FeatureType

Starts scanning features of the given segment of a sequence

Parameters:

data - the sequence

startPos - the start position of the segment

endPos - the end position of the segment

Returns:

true if successfully

hasNext

public boolean hasNext()

Description copied from interface: FeatureType

Tests if next feature exists

Returns:

true if next feature exists

next

public Feature next()

Description copied from interface: FeatureType

Gets the next feature

Returns:

the next feature if any

train

public boolean train(Dataset trainData)

Description copied from interface: FeatureType

Trains the feature types with the labeled training dataset

Specified by:

train in interface FeatureType

Overrides:

train in class AbstractFeatureType

Parameters:

trainData - the dataset for training

Returns:

true if trained successfully

readTrainingResult

public boolean readTrainingResult()

Description copied from interface: FeatureType

Reads the training result in testing mode

Specified by:

readTrainingResult in interface FeatureType

Overrides:

readTrainingResult in class AbstractFeatureType

Returns:

training result

saveTrainingResult

public boolean saveTrainingResult()

Description copied from interface: FeatureType

Saves the training results in training mode

Specified by:

saveTrainingResult in interface FeatureType

Overrides:

saveTrainingResult in class AbstractFeatureType

Returns:

true if saved successfully

supportSegment

public boolean supportSegment()

Description copied from interface: FeatureType

Tests if the feature type supports generating features for segments.

Specified by:

supportSegment in interface FeatureType

Overrides:

supportSegment in class AbstractFeatureType

Returns:

true if supported