meka.classifiers.multilabel

Class PS

java.lang.Object

weka.classifiers.AbstractClassifier

weka.classifiers.SingleClassifierEnhancer

meka.classifiers.multilabel.MultilabelClassifier

meka.classifiers.multilabel.LC

meka.classifiers.multilabel.PS

All Implemented Interfaces:

java.io.Serializable, java.lang.Cloneable, weka.classifiers.Classifier, weka.core.CapabilitiesHandler, weka.core.OptionHandler, weka.core.Randomizable, weka.core.RevisionHandler, weka.core.TechnicalInformationHandler

Direct Known Subclasses:

NSR, PSt, PSUpdateable, RAkEL

public class PS
extends LC
implements weka.core.Randomizable, weka.core.TechnicalInformationHandler

PS.java - The Pruned Sets Method. Removes examples with P-infrequent labelsets from the training data, then subsamples these labelsets N time to produce N new examples with P-frequent labelsets. Then train a standard LC classifier. The idea is to reduce the number of unique class values that would otherwise need to be learned by LC. Best used in an Ensemble (e.g., EnsembleML).
This class was rewritten and extended to be faster, use sparse LabelSets, and better OOP. There may be minor variation in results (probably on account of internal randomness, different set orderings, etc) but this should not be statistically significant.
See: Jesse Read, Bernhard Pfahringer, Geoff Holmes. Multi-label Classification using Ensembles of Pruned Sets. Proc. of IEEE International Conference on Data Mining (ICDM 2008), Pisa, Italy, 2008

Version:

April 2014

Author:

Jesse Read

See Also:

LC, Serialized Form

Constructor Summary

Constructors

Constructor and Description
PS()

Method Summary

Methods

Modifier and Type	Method and Description
void	buildClassifier(weka.core.Instances D)
int	getN() GetN - Get the subsampling value N.
java.lang.String[]	getOptions()
int	getP() GetP - Get the pruning value P.
java.lang.String	getRevision()
int	getSeed()
weka.core.TechnicalInformation	getTechnicalInformation()
java.lang.String	globalInfo() Description to display in the GUI.
java.util.Enumeration	listOptions()
static void	main(java.lang.String[] args)
void	setN(int n) SetN - Sets the subsampling value N, the (maximum) number of frequent labelsets to subsample from the infrequent labelsets.
void	setOptions(java.lang.String[] options)
void	setP(int p) SetP - Sets the pruning value P, defining an infrequent labelset as one which occurs less than P times in the data (P = 0 defaults to LC).
void	setSeed(int s) SetSeed - Use random P and N values (in this case P and N arguments determine a range of values to select from randomly, e.g., -P 1-5 selects P randomly in {1,2,3,4,5}.

Methods inherited from class meka.classifiers.multilabel.LC

distributionForInstance, toString

Methods inherited from class meka.classifiers.multilabel.MultilabelClassifier

evaluation, getCapabilities, getTemplate, makeCopies, runClassifier, testCapabilities

Methods inherited from class weka.classifiers.SingleClassifierEnhancer

classifierTipText, getClassifier, setClassifier

Methods inherited from class weka.classifiers.AbstractClassifier

classifyInstance, debugTipText, forName, getDebug, makeCopies, makeCopy, runClassifier, setDebug

Methods inherited from class java.lang.Object

equals, getClass, hashCode, notify, notifyAll, wait, wait, wait

Constructor Detail

PS

public PS()

Method Detail

globalInfo

public java.lang.String globalInfo()

Description to display in the GUI.

Overrides:

globalInfo in class LC

Returns:

the description

getTechnicalInformation

public weka.core.TechnicalInformation getTechnicalInformation()

Specified by:

getTechnicalInformation in interface weka.core.TechnicalInformationHandler

getP

public int getP()

GetP - Get the pruning value P.

setP

public void setP(int p)

SetP - Sets the pruning value P, defining an infrequent labelset as one which occurs less than P times in the data (P = 0 defaults to LC).

getN

public int getN()

GetN - Get the subsampling value N.

setN

public void setN(int n)

SetN - Sets the subsampling value N, the (maximum) number of frequent labelsets to subsample from the infrequent labelsets.

setSeed

public void setSeed(int s)

SetSeed - Use random P and N values (in this case P and N arguments determine a range of values to select from randomly, e.g., -P 1-5 selects P randomly in {1,2,3,4,5}.

Specified by:

setSeed in interface weka.core.Randomizable

getSeed

public int getSeed()

Specified by:

getSeed in interface weka.core.Randomizable

listOptions

public java.util.Enumeration listOptions()

Specified by:

listOptions in interface weka.core.OptionHandler

Overrides:

listOptions in class weka.classifiers.SingleClassifierEnhancer

setOptions

public void setOptions(java.lang.String[] options)
                throws java.lang.Exception

Specified by:

setOptions in interface weka.core.OptionHandler

Overrides:

setOptions in class weka.classifiers.SingleClassifierEnhancer

Throws:

java.lang.Exception

getOptions

public java.lang.String[] getOptions()

Specified by:

getOptions in interface weka.core.OptionHandler

Overrides:

getOptions in class weka.classifiers.SingleClassifierEnhancer

buildClassifier

public void buildClassifier(weka.core.Instances D)
                     throws java.lang.Exception

Specified by:

buildClassifier in interface weka.classifiers.Classifier

Overrides:

buildClassifier in class LC

Throws:

java.lang.Exception

getRevision

public java.lang.String getRevision()

Specified by:

getRevision in interface weka.core.RevisionHandler

Overrides:

getRevision in class LC

main

public static void main(java.lang.String[] args)