probabilityMachine
Class VennProbabilityClassifier

java.lang.Object
  |
  +--weka.classifiers.Classifier
        |
        +--probabilityMachine.VennProbabilityClassifier

All Implemented Interfaces:

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

Direct Known Subclasses:

VPMBartsRMI, VPMBartsRMI2, VPMDistMetaLearner, VPMKNearestNeighbours, VPMMDLEntropyTypeDistMetaLearner, VPMNaiveBayes, VPMSimpleTypeDistMetaLearner

public abstract class VennProbabilityClassifier

extends Classifier

Abstract classification model that produces (for each test instance) a valid probability estimation of the membership in each class.

Version:

$Revision: 1.0$

Author:

David Lindsay (davidl@cs.rhul.ac.uk)

See Also:

Serialized Form

Constructor Summary

VennProbabilityClassifier()

Method Summary

double	classifyInstance(Instance instance) Classifies the given test instance.
static void	computeRowOfVPMMatrix(java.lang.String[] typeDefinitions, Matrix vpmMatrix, Instances trainingSet, Instance testExample) Counts the number of examples with the same type as the new example to create each row of the VPM matrix.
double[]	distributionForInstance(Instance instance) Returns a single distribution for an instance, by crudely averaging the Venn probabilities.
java.lang.String	printArray(double[] array) Debugging function
java.lang.String	printArray(int[] array) Debugging function
static double[]	returnDistribution(Matrix vennProbs) Converts the Venn Probability Matrix into a distribution for each class
static double[]	returnUpperAndLowerProbability(Matrix vennProbs) Calculates the upper and lower probabilities for the predicted label (a bit like a region prediction)
abstract Matrix	vennProbsForInstance(Instance instance) Predicts the Venn probabilities for each instance.

Methods inherited from class weka.classifiers.Classifier

buildClassifier, debugTipText, forName, getDebug, getOptions, listOptions, makeCopies, setDebug, setOptions

Methods inherited from class java.lang.Object

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

Constructor Detail

VennProbabilityClassifier

public VennProbabilityClassifier()

Method Detail

vennProbsForInstance

public abstract Matrix vennProbsForInstance(Instance instance)
                                     throws java.lang.Exception

Predicts the Venn probabilities for each instance.

Parameters:

instance - the instance to be classified

Returns:

a |Y|x|Y| size array that gives the probability for each tentative classification tried.

Throws:

java.lang.Exception - if distribution could not be computed successfully

classifyInstance

public double classifyInstance(Instance instance)
                        throws java.lang.Exception

Classifies the given test instance. The instance has to belong to a dataset when it's being classified.

Overrides:

classifyInstance in class Classifier

Parameters:

instance - the instance to be classified

Returns:

the predicted most likely class for the instance or Instance.missingValue() if no prediction is made

Throws:

java.lang.Exception - if an error occurred during the prediction

distributionForInstance

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

Returns a single distribution for an instance, by crudely averaging the Venn probabilities.

Overrides:

distributionForInstance in class Classifier

Parameters:

instance - the instance to be classified

Returns:

a probability estimate for each classification

Throws:

java.lang.Exception - if an error occurred during the prediction

returnDistribution

public static double[] returnDistribution(Matrix vennProbs)
                                   throws java.lang.Exception

Converts the Venn Probability Matrix into a distribution for each class

Parameters:

vennProbs - matrix containing the Venn probabilities

Returns:

the distribution for each class

java.lang.Exception

returnUpperAndLowerProbability

public static double[] returnUpperAndLowerProbability(Matrix vennProbs)

Calculates the upper and lower probabilities for the predicted label (a bit like a region prediction)

Parameters:

vennProbs - matrix containing the Venn probabilities

Returns:

the upper and lower probability for the instance

computeRowOfVPMMatrix

public static void computeRowOfVPMMatrix(java.lang.String[] typeDefinitions,
                                         Matrix vpmMatrix,
                                         Instances trainingSet,
                                         Instance testExample)
                                  throws java.lang.Exception

Counts the number of examples with the same type as the new example to create each row of the VPM matrix.

Parameters:

typeDefinitions - this contains the training types + new test example at the end.

vpmMatrix - the current VPM matrix, this will be updated in the function

java.lang.Exception

printArray

public java.lang.String printArray(double[] array)

Debugging function

printArray

public java.lang.String printArray(int[] array)

Debugging function