mulan.classifier.lazy
Class BRkNN

java.lang.Object
  mulan.classifier.MultiLabelLearnerBase
      mulan.classifier.lazy.MultiLabelKNN
          mulan.classifier.lazy.BRkNN

All Implemented Interfaces:

Serializable, MultiLabelLearner, TechnicalInformationHandler

public class BRkNN
extends MultiLabelKNN

Simple BR implementation of the KNN algorithm.For more information, see

Eleftherios Spyromitros, Grigorios Tsoumakas, Ioannis Vlahavas: An Empirical Study of Lazy Multilabel Classification Algorithms. In: Proc. 5th Hellenic Conference on Artificial Intelligence (SETN 2008), 2008.

BibTeX:

 @inproceedings{EleftheriosSpyromitros2008,
    author = {Eleftherios Spyromitros, Grigorios Tsoumakas, Ioannis Vlahavas},
    booktitle = {Proc. 5th Hellenic Conference on Artificial Intelligence (SETN 2008)},
    title = {An Empirical Study of Lazy Multilabel Classification Algorithms},
    year = {2008},
    location = {Syros, Greece}
 }
 

Version:

2010.12.29

Author:

Eleftherios Spyromitros-Xioufis, Grigorios Tsoumakas

See Also:

Serialized Form

Nested Class Summary

static class

BRkNN.ExtensionType The two types of extensions

Field Summary

Fields inherited from class mulan.classifier.lazy.MultiLabelKNN

dfunc, distanceWeighting, lnn, numOfNeighbors, train, WEIGHT_INVERSE, WEIGHT_NONE, WEIGHT_SIMILARITY

Fields inherited from class mulan.classifier.MultiLabelLearnerBase

featureIndices, labelIndices, numLabels

Constructor Summary

BRkNN()
Default constructor

BRkNN(int numOfNeighbors)
A constructor that sets the number of neighbors

BRkNN(int numOfNeighbors, BRkNN.ExtensionType ext)
Constructor giving the option to select an extension of the base version

Method Summary

protected void	buildInternal(MultiLabelInstances aTrain) Learner specific implementation of building the model from MultiLabelInstances training data set.
TechnicalInformation	getTechnicalInformation() Returns an instance of a TechnicalInformation object, containing detailed information about the technical background of this class, e.g., paper reference or book this class is based on.
String	globalInfo() Returns a string describing the multi-label learner.
protected boolean[]	labelsFromConfidences2(double[] confidences) used for BRknn-a
protected boolean[]	labelsFromConfidences3(double[] confidences) used for BRkNN-b (break ties arbitrarily)
protected MultiLabelOutput	makePredictionInternal(Instance instance) weka Ibk style prediction
void	setCvMaxK(int cvMaxK) set the maximum number of neighbors to be evaluated via cross-validation
void	setkSelectionViaCV(boolean flag)

Methods inherited from class mulan.classifier.lazy.MultiLabelKNN

isUpdatable, setDfunc, setDistanceWeighting

Methods inherited from class mulan.classifier.MultiLabelLearnerBase

build, debug, getDebug, isModelInitialized, makeCopy, makePrediction, setDebug

Methods inherited from class java.lang.Object

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

Constructor Detail

BRkNN

public BRkNN()

Default constructor

BRkNN

public BRkNN(int numOfNeighbors)

A constructor that sets the number of neighbors

Parameters:

numOfNeighbors -

BRkNN

public BRkNN(int numOfNeighbors,
             BRkNN.ExtensionType ext)

Constructor giving the option to select an extension of the base version

Parameters:

numOfNeighbors -

ext - the extension to use (see BRkNN.ExtensionType)

Method Detail

getTechnicalInformation

public TechnicalInformation getTechnicalInformation()

Description copied from class: MultiLabelLearnerBase

Returns an instance of a TechnicalInformation object, containing detailed information about the technical background of this class, e.g., paper reference or book this class is based on.

Specified by:

getTechnicalInformation in interface TechnicalInformationHandler

Specified by:

getTechnicalInformation in class MultiLabelLearnerBase

Returns:

the technical information about this class

buildInternal

protected void buildInternal(MultiLabelInstances aTrain)
                      throws Exception

Description copied from class: MultiLabelLearnerBase

Learner specific implementation of building the model from MultiLabelInstances training data set. This method is called from MultiLabelLearnerBase.build(MultiLabelInstances) method, where behavior common across all learners is applied.

Overrides:

buildInternal in class MultiLabelKNN

Parameters:

aTrain - the training data set

Throws:

Exception - if learner model was not created successfully

setkSelectionViaCV

public void setkSelectionViaCV(boolean flag)

Parameters:

flag - if true the k is selected via cross-validation

makePredictionInternal

protected MultiLabelOutput makePredictionInternal(Instance instance)
                                           throws Exception

weka Ibk style prediction

Specified by:

makePredictionInternal in class MultiLabelLearnerBase

Parameters:

instance - the data instance to predict on

Returns:

the output of the learner for the given instance

Throws:

Exception - if nearest neighbours search fails

labelsFromConfidences2

protected boolean[] labelsFromConfidences2(double[] confidences)

used for BRknn-a

Parameters:

confidences - the probabilities for each label

Returns:

a bipartition

labelsFromConfidences3

protected boolean[] labelsFromConfidences3(double[] confidences)

used for BRkNN-b (break ties arbitrarily)

Parameters:

confidences - the probabilities for each label

Returns:

a bipartition

setCvMaxK

public void setCvMaxK(int cvMaxK)

set the maximum number of neighbors to be evaluated via cross-validation

Parameters:

cvMaxK -

globalInfo

public String globalInfo()

Description copied from class: MultiLabelLearnerBase

Returns a string describing the multi-label learner.

Specified by:

globalInfo in class MultiLabelLearnerBase

Returns:

a description suitable for displaying in a future gui