eus.ixa.ixa.pipe.ml.document

Class DocumentClassifierME

java.lang.Object

eus.ixa.ixa.pipe.ml.document.DocumentClassifierME

All Implemented Interfaces:

DocumentClassifier

public class DocumentClassifierME
extends Object
implements DocumentClassifier

Statistical implementation of DocumentClassifier.

Constructor Summary

Constructors

Constructor and Description
DocumentClassifierME(DocumentClassifierModel model) Initializes the current instance with a DocumentClassifier model.

Method Summary

Modifier and Type	Method and Description
String	classify(String[] text) Classifies the given text, provided in separate tokens.
double[]	classifyProb(String[] text) Classify the given text provided as tokens.
void	clearFeatureData() Forgets all adaptive data which was collected during previous calls to one of the find methods.
String	getAllLabels(double[] results) Get the label associated with the given probabilities.
String	getBestLabel(double[] outcome) Get the best label from previously generated outcome probabilities.
int	getIndex(String category) Get the index of a certain label.
String	getLabel(int index) Get the category at a given index.
int	getNumberOfLabels() get the number of found labels.
Map<String,Double>	scoreMap(String[] text) Returns a map in which the key is the label name and the value is the score.
SortedMap<Double,Set<String>>	sortedScoreMap(String[] text) Returns a map with the score as a key in ascending order.
static DocumentClassifierModel	train(String languageCode, opennlp.tools.util.ObjectStream<DocSample> samples, opennlp.tools.util.TrainingParameters mlParams, DocumentClassifierFactory factory)

Methods inherited from class java.lang.Object

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

Constructor Detail

DocumentClassifierME

public DocumentClassifierME(DocumentClassifierModel model)

Initializes the current instance with a DocumentClassifier model. Default feature generation is used.

Parameters:

model - the document classifier model

Method Detail

classify

public String classify(String[] text)

Description copied from interface: DocumentClassifier

Classifies the given text, provided in separate tokens.

Specified by:

classify in interface DocumentClassifier

Parameters:

text - the tokens of text to classify

Returns:

the best label found

classifyProb

public double[] classifyProb(String[] text)

Classify the given text provided as tokens.

Specified by:

classifyProb in interface DocumentClassifier

Parameters:

text - text tokens to classify

Returns:

per category probabilities

getBestLabel

public String getBestLabel(double[] outcome)

Description copied from interface: DocumentClassifier

Get the best label from previously generated outcome probabilities.

Specified by:

getBestLabel in interface DocumentClassifier

Parameters:

outcome - a vector of outcome probabilities

Returns:

the best label in a String

scoreMap

public Map<String,Double> scoreMap(String[] text)

Returns a map in which the key is the label name and the value is the score.

Specified by:

scoreMap in interface DocumentClassifier

Parameters:

text - the input text to classify

Returns:

the score map

sortedScoreMap

public SortedMap<Double,Set<String>> sortedScoreMap(String[] text)

Returns a map with the score as a key in ascending order. The value is a Set of categories with the score. Many labels can have the same score, hence the Set as value.

Specified by:

sortedScoreMap in interface DocumentClassifier

Parameters:

text - the input text to classify

Returns:

the sorted score map

getIndex

public int getIndex(String category)

Description copied from interface: DocumentClassifier

Get the index of a certain label.

Specified by:

getIndex in interface DocumentClassifier

Parameters:

category - the label

Returns:

an index

getLabel

public String getLabel(int index)

Description copied from interface: DocumentClassifier

Get the category at a given index.

Specified by:

getLabel in interface DocumentClassifier

Parameters:

index - the index

Returns:

a label

getNumberOfLabels

public int getNumberOfLabels()

Description copied from interface: DocumentClassifier

get the number of found labels.

Specified by:

getNumberOfLabels in interface DocumentClassifier

Returns:

the number of labels

getAllLabels

public String getAllLabels(double[] results)

Description copied from interface: DocumentClassifier

Get the label associated with the given probabilities.

Specified by:

getAllLabels in interface DocumentClassifier

Parameters:

results - the probabilities of each label

Returns:

the name of the outcome

clearFeatureData

public void clearFeatureData()

Forgets all adaptive data which was collected during previous calls to one of the find methods. This method is typical called at the end of a document.

Specified by:

clearFeatureData in interface DocumentClassifier

train

public static DocumentClassifierModel train(String languageCode,
                                            opennlp.tools.util.ObjectStream<DocSample> samples,
                                            opennlp.tools.util.TrainingParameters mlParams,
                                            DocumentClassifierFactory factory)
                                     throws IOException

Throws:

IOException