confidenceMachine.tcm
Class TCMKNearestNeighbours

java.lang.Object
  |
  +--weka.classifiers.Classifier
        |
        +--confidenceMachine.ConfidenceClassifier
              |
              +--confidenceMachine.tcm.TCMKNearestNeighbours

All Implemented Interfaces:

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

public class TCMKNearestNeighbours

extends ConfidenceClassifier

implements UpdateableClassifier, OptionHandler

The TCM K-Nearest Neighbours algorithm. With this particular implementation is embedded, meaning that the traditional K has been altered slightly to adapt a strangeness measure using a ratio of the K nearest distances to the predicted class and all other classes. For more information, see

Transductive Confidence Machines for Pattern Recognition, CRLC Technical Report CRLC-TR-01-02, June 18 2001, Kostas Proedrou, Ilia Nouretdinov, Volodya Vovk, Alex Gammerman -Q Distance Metric
This is the distance metric that we are using (default Euclidean distance)

Version:

$Revision: 1.0 $

Author:

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

See Also:

Serialized Form

Constructor Summary

TCMKNearestNeighbours()
The amazing TCM K Nearest Neighbours classifier

TCMKNearestNeighbours(int k)
The amazing TCM K Nearest Neighbours classifier

Method Summary

void	buildClassifier(Instances instances) Generates the classifier.
int	getDebugEvery() Reports whether the TCM is currently in debug mode (if not -1), and specifies how frequently to output message about TCM's progress on data.
java.lang.String	getDistanceMetric() The gives details about the distance metric that has been chosen
int	getKNN() Gets the number of neighbours the learner will use.
java.lang.String[]	getOptions() Gets the current settings of TCM.
java.util.Enumeration	listOptions() Returns an enumeration describing the available options
static void	main(java.lang.String[] argv) Main method for testing this class.
double[]	pValuesForInstance(Instance instance) Returns the p-values for a given test instance.
void	setDebugEvery(int debug) Set the TCM to debug its output, outputting progress as it goes along.
boolean	setDistanceMetric(java.lang.String distName, java.lang.String[] distOptions) Set the distance metric
void	setKNN(int k) Set the number of neighbours the learner is to use.
void	setOptions(java.lang.String[] options) Parses a given list of options.
java.lang.String	toString() Returns a description of this classifier.
void	updateClassifier(Instance instance) Updates the classifier.

Methods inherited from class confidenceMachine.ConfidenceClassifier

calculateConfidenceAndCredibility, calculatePValue, calculatePValue, classifyInstance, returnRegionPrediction

Methods inherited from class weka.classifiers.Classifier

debugTipText, distributionForInstance, forName, getDebug, makeCopies, setDebug

Methods inherited from class java.lang.Object

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

Constructor Detail

TCMKNearestNeighbours

public TCMKNearestNeighbours(int k)

The amazing TCM K Nearest Neighbours classifier

Parameters:

k - the number of nearest neighbors to use for prediction

TCMKNearestNeighbours

public TCMKNearestNeighbours()

The amazing TCM K Nearest Neighbours classifier

Method Detail

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

Updates the classifier.

Specified by:

updateClassifier in interface UpdateableClassifier

Parameters:

instance - the instance to be put into the classifier

Throws:

java.lang.Exception - if the instance could not be included successfully

setKNN

public void setKNN(int k)

Set the number of neighbours the learner is to use.

Parameters:

k - the number of neighbours.

getKNN

public int getKNN()

Gets the number of neighbours the learner will use.

Returns:

the number of neighbours.

setDebugEvery

public void setDebugEvery(int debug)

Set the TCM to debug its output, outputting progress as it goes along. (Maybe also output alphas?) This is very useful when processing very large data sets, to keep track of its progress (defualt: -1) which specifies no output at all.

getDebugEvery

public int getDebugEvery()

Reports whether the TCM is currently in debug mode (if not -1), and specifies how frequently to output message about TCM's progress on data.

Returns:

how fre

setDistanceMetric

public boolean setDistanceMetric(java.lang.String distName,
                                 java.lang.String[] distOptions)

Set the distance metric

Parameters:

distName - the distance metric class name in full

distOptions - the options specified for that distance metric

Returns:

true if successful

getDistanceMetric

public java.lang.String getDistanceMetric()

The gives details about the distance metric that has been chosen

Returns:

true if default Euclidean chosen, false other wise

listOptions

public java.util.Enumeration listOptions()

Returns an enumeration describing the available options

Specified by:

listOptions in interface OptionHandler

Overrides:

listOptions in class Classifier

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:

-K num
Set the number of nearest neighbors to use in prediction (default 1)

Specified by:

setOptions in interface OptionHandler

Overrides:

setOptions in class Classifier

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 TCM.

Specified by:

getOptions in interface OptionHandler

Overrides:

getOptions in class Classifier

Returns:

an array of strings suitable for passing to setOptions()

pValuesForInstance

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

Returns the p-values for a given test instance.

Specified by:

pValuesForInstance in class ConfidenceClassifier

Parameters:

instance - the instance to be classified

Returns:

the probabilities for each class (normalised p-values)

Throws:

java.lang.Exception - no training instances

toString

public java.lang.String toString()

Returns a description of this classifier.

Overrides:

toString in class java.lang.Object

Returns:

a description of this classifier as a string.

main

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

Main method for testing this class.

Parameters:

argv - should contain command line arguments for evaluation (see Evaluation).