weka.classifiers.lazy
Class LWL

java.lang.Object
  weka.classifiers.Classifier
      weka.classifiers.SingleClassifierEnhancer
          weka.classifiers.lazy.LWL

All Implemented Interfaces:

java.lang.Cloneable, OptionHandler, java.io.Serializable, UpdateableClassifier, WeightedInstancesHandler

public class LWL

extends SingleClassifierEnhancer

implements UpdateableClassifier, WeightedInstancesHandler

Locally-weighted learning. Uses an instance-based algorithm to assign instance weights which are then used by a specified WeightedInstancesHandler. A good choice for classification is NaiveBayes. LinearRegression is suitable for regression problems. For more information, see

Eibe Frank, Mark Hall, and Bernhard Pfahringer (2003). Locally Weighted Naive Bayes. Working Paper 04/03, Department of Computer Science, University of Waikato. Atkeson, C., A. Moore, and S. Schaal (1996) Locally weighted learning download postscript.

Valid options are:

-D
Produce debugging output.

-K num
Set the number of neighbours used for setting kernel bandwidth. (default all)

-W num
Set the weighting kernel shape to use. 1 = Inverse, 2 = Gaussian. (default 0 = Linear)

-B classname
Specify the full class name of a base classifier (which needs to be a WeightedInstancesHandler).

Version:

$Revision: 1.9 $

Author:

Len Trigg (trigg@cs.waikato.ac.nz), Eibe Frank (eibe@cs.waikato.ac.nz)

See Also:

Serialized Form

Field Summary

protected static int	GAUSS
protected static int	INVERSE
protected static int	LINEAR The available kernel weighting methods
protected int	m_kNN The number of neighbours used to select the kernel bandwidth
protected double[]	m_Max The maximum values for numeric attributes.
protected double[]	m_Min The minimum values for numeric attributes.
protected Instances	m_Train The training instances used for classification.
protected boolean	m_UseAllK True if m_kNN should be set to all instances
protected int	m_WeightKernel The weighting kernel method currently selected

Fields inherited from class weka.classifiers.SingleClassifierEnhancer

m_Classifier

Fields inherited from class weka.classifiers.Classifier

m_Debug

Constructor Summary

LWL()
Constructor.

Method Summary

void	buildClassifier(Instances instances) Generates the classifier.
protected java.lang.String	defaultClassifierString() String describing default classifier.
private double	difference(int index, double val1, double val2) Computes the difference between two given attribute values.
private double	distance(Instance first, Instance second) Calculates the distance between two instances
double[]	distributionForInstance(Instance instance) Calculates the class membership probabilities for the given test instance.
protected double	getAttributeMax(int index) Gets an attributes maximum observed value
protected double	getAttributeMin(int index) Gets an attributes minimum observed value
int	getKNN() Gets the number of neighbours used for kernel bandwidth setting.
java.lang.String[]	getOptions() Gets the current settings of the classifier.
int	getWeightingKernel() Gets the kernel weighting method to use.
java.lang.String	globalInfo() Returns a string describing classifier
java.lang.String	KNNTipText() Returns the tip text for this property
java.util.Enumeration	listOptions() Returns an enumeration describing the available options.
static void	main(java.lang.String[] argv) Main method for testing this class.
private double	norm(double x, int i) Normalizes a given value of a numeric attribute.
void	setKNN(int knn) Sets the number of neighbours used for kernel bandwidth setting.
void	setOptions(java.lang.String[] options) Parses a given list of options.
void	setWeightingKernel(int kernel) Sets the kernel weighting method to use.
java.lang.String	toString() Returns a description of this classifier.
void	updateClassifier(Instance instance) Adds the supplied instance to the training set
private void	updateMinMax(Instance instance) Updates the minimum and maximum values for all the attributes based on a new instance.
java.lang.String	weightingKernelTipText() Returns the tip text for this property

Methods inherited from class weka.classifiers.SingleClassifierEnhancer

classifierTipText, getClassifier, getClassifierSpec, setClassifier

Methods inherited from class weka.classifiers.Classifier

classifyInstance, debugTipText, forName, getDebug, makeCopies, setDebug

Methods inherited from class java.lang.Object

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

Field Detail

m_Train

protected Instances m_Train

The training instances used for classification.

m_Min

protected double[] m_Min

The minimum values for numeric attributes.

m_Max

protected double[] m_Max

The maximum values for numeric attributes.

m_kNN

protected int m_kNN

The number of neighbours used to select the kernel bandwidth

m_WeightKernel

protected int m_WeightKernel

The weighting kernel method currently selected

m_UseAllK

protected boolean m_UseAllK

True if m_kNN should be set to all instances

LINEAR

protected static final int LINEAR

The available kernel weighting methods

See Also:

Constant Field Values

INVERSE

protected static final int INVERSE

See Also:

Constant Field Values

GAUSS

protected static final int GAUSS

See Also:

Constant Field Values

Constructor Detail

LWL

public LWL()

Constructor.

Method Detail

globalInfo

public java.lang.String globalInfo()

Returns a string describing classifier

Returns:

a description suitable for displaying in the explorer/experimenter gui

defaultClassifierString

protected java.lang.String defaultClassifierString()

String describing default classifier.

Overrides:

defaultClassifierString in class SingleClassifierEnhancer

listOptions

public java.util.Enumeration listOptions()

Returns an enumeration describing the available options.

Specified by:

listOptions in interface OptionHandler

Overrides:

listOptions in class SingleClassifierEnhancer

Returns:

an enumeration of all the available options.

setOptions

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

Parses a given list of options. Valid options are:

-D
Produce debugging output.

-K num
Set the number of neighbours used for setting kernel bandwidth. (default all)

-W num
Set the weighting kernel shape to use. 1 = Inverse, 2 = Gaussian. (default 0 = Linear)

-B classname
Specify the full class name of a base classifier (which needs to be a WeightedInstancesHandler).

Specified by:

setOptions in interface OptionHandler

Overrides:

setOptions in class SingleClassifierEnhancer

Parameters:

options - the list of options as an array of strings

Throws:

java.lang.Exception - if an option is not supported

getOptions

public java.lang.String[] getOptions()

Gets the current settings of the classifier.

Specified by:

getOptions in interface OptionHandler

Overrides:

getOptions in class SingleClassifierEnhancer

Returns:

an array of strings suitable for passing to setOptions

KNNTipText

public java.lang.String KNNTipText()

Returns the tip text for this property

Returns:

tip text for this property suitable for displaying in the explorer/experimenter gui

setKNN

public void setKNN(int knn)

Sets the number of neighbours used for kernel bandwidth setting. The bandwidth is taken as the distance to the kth neighbour.

Parameters:

knn - the number of neighbours included inside the kernel bandwidth, or 0 to specify using all neighbors.

getKNN

public int getKNN()

Gets the number of neighbours used for kernel bandwidth setting. The bandwidth is taken as the distance to the kth neighbour.

Returns:

the number of neighbours included inside the kernel bandwidth, or 0 for all neighbours

weightingKernelTipText

public java.lang.String weightingKernelTipText()

Returns the tip text for this property

Returns:

tip text for this property suitable for displaying in the explorer/experimenter gui

setWeightingKernel

public void setWeightingKernel(int kernel)

Sets the kernel weighting method to use. Must be one of LINEAR, INVERSE, or GAUSS, other values are ignored.

Parameters:

kernel - the new kernel method to use. Must be one of LINEAR, INVERSE, or GAUSS

getWeightingKernel

public int getWeightingKernel()

Gets the kernel weighting method to use.

Returns:

the new kernel method to use. Will be one of LINEAR, INVERSE, or GAUSS

getAttributeMin

protected double getAttributeMin(int index)

Gets an attributes minimum observed value

Parameters:

index - the index of the attribute

Returns:

the minimum observed value

getAttributeMax

protected double getAttributeMax(int index)

Gets an attributes maximum observed value

Parameters:

index - the index of the attribute

Returns:

the maximum observed value

buildClassifier

public void buildClassifier(Instances instances)
                     throws java.lang.Exception

Generates the classifier.

Specified by:

buildClassifier in class Classifier

Parameters:

instances - set of instances serving as training data

Throws:

java.lang.Exception - if the classifier has not been generated successfully

updateClassifier

public void updateClassifier(Instance instance)
                      throws java.lang.Exception

Adds the supplied instance to the training set

Specified by:

updateClassifier in interface UpdateableClassifier

Parameters:

instance - the instance to add

Throws:

java.lang.Exception - if instance could not be incorporated successfully

distributionForInstance

public double[] distributionForInstance(Instance instance)
                                 throws java.lang.Exception

Calculates the class membership probabilities for the given test instance.

Overrides:

distributionForInstance in class Classifier

Parameters:

instance - the instance to be classified

Returns:

preedicted class probability distribution

Throws:

java.lang.Exception - if distribution can't be computed successfully

toString

public java.lang.String toString()

Returns a description of this classifier.

Returns:

a description of this classifier as a string.

distance

private double distance(Instance first,
                        Instance second)

Calculates the distance between two instances

Returns:

the distance between the two given instances, between 0 and 1

difference

private double difference(int index,
                          double val1,
                          double val2)

Computes the difference between two given attribute values.

norm

private double norm(double x,
                    int i)

Normalizes a given value of a numeric attribute.

Parameters:

x - the value to be normalized

i - the attribute's index

updateMinMax

private void updateMinMax(Instance instance)

Updates the minimum and maximum values for all the attributes based on a new instance.

Parameters:

instance - the new instance

main

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

Main method for testing this class.

Parameters:

argv - the options