weka.classifiers.functions.pace

Class MixtureDistribution

java.lang.Object

weka.classifiers.functions.pace.MixtureDistribution

All Implemented Interfaces:

RevisionHandler, TechnicalInformationHandler

Direct Known Subclasses:

ChisqMixture, NormalMixture

public abstract class MixtureDistribution
extends java.lang.Object
implements TechnicalInformationHandler, RevisionHandler

Abtract class for manipulating mixture distributions.

REFERENCES

Wang, Y. (2000). "A new approach to fitting linear models in high dimensional spaces." PhD Thesis. Department of Computer Science, University of Waikato, New Zealand.

Wang, Y. and Witten, I. H. (2002). "Modeling for optimal probability prediction." Proceedings of ICML'2002. Sydney.

Version:

$Revision: 1.5 $

Author:

Yong Wang (yongwang@cs.waikato.ac.nz)

Field Summary

Fields

Modifier and Type	Field and Description
static int	NNMMethod The nonnegative-measure-based method
static int	PMMethod The probability-measure-based method

Constructor Summary

Constructors

Constructor and Description
MixtureDistribution()

Method Summary

Modifier and Type	Method and Description
PaceMatrix	empiricalProbability(DoubleVector data, PaceMatrix intervals) Computes the empirical probabilities of the data over a set of intervals.
void	fit(DoubleVector data) Fits the mixture (or mixing) distribution to the data.
void	fit(DoubleVector data, int method) Fits the mixture (or mixing) distribution to the data.
DiscreteFunction	fitForSingleCluster(DoubleVector data, int method) Fits the mixture (or mixing) distribution to the data.
abstract PaceMatrix	fittingIntervals(DoubleVector data) Contructs the set of fitting intervals for mixture estimation.
DiscreteFunction	getMixingDistribution() Gets the mixing distribution
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.
abstract PaceMatrix	probabilityMatrix(DoubleVector s, PaceMatrix intervals) Contructs the probability matrix for mixture estimation, given a set of support points and a set of intervals.
abstract boolean	separable(DoubleVector data, int i0, int i1, double x) Return true if a value can be considered for mixture estimatino separately from the data indexed between i0 and i1
void	setMixingDistribution(DiscreteFunction d) Sets the mixing distribution
abstract DoubleVector	supportPoints(DoubleVector data, int ne) Contructs the set of support points for mixture estimation.
java.lang.String	toString() Converts to a string

Methods inherited from class java.lang.Object

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

Methods inherited from interface weka.core.RevisionHandler

getRevision

Field Detail

NNMMethod

public static final int NNMMethod

The nonnegative-measure-based method

See Also:

Constant Field Values

PMMethod

public static final int PMMethod

The probability-measure-based method

See Also:

Constant Field Values

Constructor Detail

MixtureDistribution

public MixtureDistribution()

Method Detail

getTechnicalInformation

public 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.

Specified by:

getTechnicalInformation in interface TechnicalInformationHandler

Returns:

the technical information about this class

getMixingDistribution

public DiscreteFunction getMixingDistribution()

Gets the mixing distribution

Returns:

the mixing distribution

setMixingDistribution

public void setMixingDistribution(DiscreteFunction d)

Sets the mixing distribution

Parameters:

d - the mixing distribution

fit

public void fit(DoubleVector data)

Fits the mixture (or mixing) distribution to the data. The default method is the nonnegative-measure-based method.

Parameters:

data - the data, supposedly generated from the mixture model

fit

public void fit(DoubleVector data,
                int method)

Fits the mixture (or mixing) distribution to the data.

Parameters:

data - the data supposedly generated from the mixture

method - the method to be used. Refer to the static final variables of this class.

fitForSingleCluster

public DiscreteFunction fitForSingleCluster(DoubleVector data,
                                            int method)

Fits the mixture (or mixing) distribution to the data. The data is not pre-clustered for computational efficiency.

Parameters:

data - the data supposedly generated from the mixture

method - the method to be used. Refer to the static final variables of this class.

Returns:

the generated distribution

separable

public abstract boolean separable(DoubleVector data,
                                  int i0,
                                  int i1,
                                  double x)

Return true if a value can be considered for mixture estimatino separately from the data indexed between i0 and i1

Parameters:

data - the data supposedly generated from the mixture

i0 - the index of the first element in the group

i1 - the index of the last element in the group

x - the value

Returns:

true if a value can be considered

supportPoints

public abstract DoubleVector supportPoints(DoubleVector data,
                                           int ne)

Contructs the set of support points for mixture estimation.

Parameters:

data - the data supposedly generated from the mixture

ne - the number of extra data that are suppposedly discarded earlier and not passed into here

Returns:

the set of support points

fittingIntervals

public abstract PaceMatrix fittingIntervals(DoubleVector data)

Contructs the set of fitting intervals for mixture estimation.

Parameters:

data - the data supposedly generated from the mixture

Returns:

the set of fitting intervals

probabilityMatrix

public abstract PaceMatrix probabilityMatrix(DoubleVector s,
                                             PaceMatrix intervals)

Contructs the probability matrix for mixture estimation, given a set of support points and a set of intervals.

Parameters:

s - the set of support points

intervals - the intervals

Returns:

the probability matrix

empiricalProbability

public PaceMatrix empiricalProbability(DoubleVector data,
                                       PaceMatrix intervals)

Computes the empirical probabilities of the data over a set of intervals.

Parameters:

data - the data

intervals - the intervals

Returns:

the empirical probabilities

toString

public java.lang.String toString()

Converts to a string

Overrides:

toString in class java.lang.Object

Returns:

a string representation