moa.clusterers

Class CobWeb

java.lang.Object

moa.AbstractMOAObject

moa.options.AbstractOptionHandler

moa.clusterers.AbstractClusterer

moa.clusterers.CobWeb

All Implemented Interfaces:

Serializable, Clusterer, AWTRenderable, MOAObject, OptionHandler

public class CobWeb
extends AbstractClusterer

Class implementing the Cobweb and Classit clustering algorithms. See: http://en.wikipedia.org/wiki/Cobweb_%28clustering%29 Citation: D. Fisher (1987). Knowledge acquisition via incremental conceptual clustering. Machine Learning. 2(2):139-172.

See Also:

Serialized Form

Field Summary

Fields

Modifier and Type	Field and Description
FloatOption	acuityOption
static String	classifierPurposeString
FloatOption	cutoffOption
protected double	m_acuity Acuity (minimum standard deviation).
protected moa.clusterers.CobWeb.CNode	m_cobwebTree Holds the root of the Cobweb tree.
protected double	m_cutoff Cutoff (minimum category utility).
protected static double	m_normal Normal constant.
protected int	m_numberMerges the number of merges that happened
protected int	m_numberOfClusters Number of clusters (nodes in the tree).
protected boolean	m_numberOfClustersDetermined whether the number of clusters was already determined
protected int	m_numberSplits the number of splits that happened
protected boolean	m_saveInstances Output instances in graph representation of Cobweb tree (Allows instances at nodes in the tree to be visualized in the Explorer).
IntOption	randomSeedOption

Fields inherited from class moa.clusterers.AbstractClusterer

clustererRandom, clustering, evaluateMicroClusteringOption, modelContext, randomSeed, trainingWeightSeenByModel

Fields inherited from class moa.options.AbstractOptionHandler

classOptionNamesToPreparedObjects, options

Constructor Summary

Constructors

Constructor and Description
CobWeb()

Method Summary

Methods

Modifier and Type	Method and Description
protected void	determineNumberOfClusters() determines the number of clusters if necessary
double	getAcuity() get the acuity value
Clustering	getClusteringResult()
double	getCutoff() get the cutoff
void	getModelDescription(StringBuilder out, int indent)
protected Measurement[]	getModelMeasurementsImpl()
boolean	getSaveInstanceData() Get the value of saveInstances.
double[]	getVotesForInstance(weka.core.Instance instance) Classifies a given instance.
String	graph() Generates the graph string of the Cobweb tree
boolean	isRandomizable()
int	numberOfClusters() Returns the number of clusters.
void	resetLearningImpl()
void	setAcuity(double a) set the acuity.
void	setCutoff(double c) set the cutoff
void	setSaveInstanceData(boolean newsaveInstances) Set the value of saveInstances.
void	trainOnInstanceImpl(weka.core.Instance newInstance) Adds an instance to the clusterer.

Methods inherited from class moa.clusterers.AbstractClusterer

contextIsCompatible, copy, getAttributeNameString, getAWTRenderer, getClassLabelString, getClassNameString, getDescription, getMicroClusteringResult, getModelContext, getModelMeasurements, getNominalValueString, getPurposeString, getSubClusterers, implementsMicroClusterer, keepClassLabel, modelAttIndexToInstanceAttIndex, modelAttIndexToInstanceAttIndex, prepareForUseImpl, resetLearning, setModelContext, setRandomSeed, trainingHasStarted, trainingWeightSeenByModel, trainOnInstance

Methods inherited from class moa.options.AbstractOptionHandler

discoverOptionsViaReflection, getCLICreationString, getOptions, getPreparedClassOption, getPreparedClassOption, prepareClassOptions, prepareForUse, prepareForUse

Methods inherited from class moa.AbstractMOAObject

copy, measureByteSize, measureByteSize, toString

Methods inherited from class java.lang.Object

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

Methods inherited from interface moa.options.OptionHandler

getCLICreationString, getOptions, prepareForUse, prepareForUse

Methods inherited from interface moa.MOAObject

measureByteSize

Field Detail

acuityOption

public FloatOption acuityOption

cutoffOption

public FloatOption cutoffOption

randomSeedOption

public IntOption randomSeedOption

m_normal

protected static final double m_normal

Normal constant.

m_acuity

protected double m_acuity

Acuity (minimum standard deviation).

m_cutoff

protected double m_cutoff

Cutoff (minimum category utility).

m_cobwebTree

protected moa.clusterers.CobWeb.CNode m_cobwebTree

Holds the root of the Cobweb tree.

m_numberOfClusters

protected int m_numberOfClusters

Number of clusters (nodes in the tree). Must never be queried directly, only via the method numberOfClusters(). Otherwise it's not guaranteed that it contains the correct value.

See Also:

numberOfClusters(), m_numberOfClustersDetermined

m_numberOfClustersDetermined

protected boolean m_numberOfClustersDetermined

whether the number of clusters was already determined

m_numberSplits

protected int m_numberSplits

the number of splits that happened

m_numberMerges

protected int m_numberMerges

the number of merges that happened

m_saveInstances

protected boolean m_saveInstances

Output instances in graph representation of Cobweb tree (Allows instances at nodes in the tree to be visualized in the Explorer).

classifierPurposeString

public static final String classifierPurposeString

See Also:

Constant Field Values

Constructor Detail

CobWeb

public CobWeb()

Method Detail

resetLearningImpl

public void resetLearningImpl()

Specified by:

resetLearningImpl in class AbstractClusterer

trainOnInstanceImpl

public void trainOnInstanceImpl(weka.core.Instance newInstance)

Adds an instance to the clusterer.

Specified by:

trainOnInstanceImpl in class AbstractClusterer

Parameters:

newInstance - the instance to be added

Throws:

Exception - if something goes wrong

getVotesForInstance

public double[] getVotesForInstance(weka.core.Instance instance)

Classifies a given instance.

Parameters:

instance - the instance to be assigned to a cluster

Returns:

the number of the assigned cluster as an interger if the class is enumerated, otherwise the predicted value

Throws:

Exception - if instance could not be classified successfully

determineNumberOfClusters

protected void determineNumberOfClusters()

determines the number of clusters if necessary

See Also:

m_numberOfClusters, m_numberOfClustersDetermined

numberOfClusters

public int numberOfClusters()

Returns the number of clusters.

Returns:

the number of clusters

getModelMeasurementsImpl

protected Measurement[] getModelMeasurementsImpl()

Specified by:

getModelMeasurementsImpl in class AbstractClusterer

getModelDescription

public void getModelDescription(StringBuilder out,
                       int indent)

Specified by:

getModelDescription in class AbstractClusterer

isRandomizable

public boolean isRandomizable()

graph

public String graph()

Generates the graph string of the Cobweb tree

Returns:

a String value

Throws:

Exception - if an error occurs

setAcuity

public void setAcuity(double a)

set the acuity.

Parameters:

a - the acuity value

getAcuity

public double getAcuity()

get the acuity value

Returns:

the acuity

setCutoff

public void setCutoff(double c)

set the cutoff

Parameters:

c - the cutof

getCutoff

public double getCutoff()

get the cutoff

Returns:

the cutoff

getSaveInstanceData

public boolean getSaveInstanceData()

Get the value of saveInstances.

Returns:

Value of saveInstances.

setSaveInstanceData

public void setSaveInstanceData(boolean newsaveInstances)

Set the value of saveInstances.

Parameters:

newsaveInstances - Value to assign to saveInstances.

getClusteringResult

public Clustering getClusteringResult()