moa.classifiers.rules

Class RuleClassifier

java.lang.Object

moa.AbstractMOAObject

moa.options.AbstractOptionHandler

moa.classifiers.AbstractClassifier

moa.classifiers.rules.RuleClassifier

All Implemented Interfaces:

Serializable, Classifier, AWTRenderable, MOAObject, OptionHandler

Direct Known Subclasses:

RuleClassifierNBayes

public class RuleClassifier
extends AbstractClassifier

This classifier learn ordered and unordered rule set from data stream. This algorithm also does the detection of anomalies.

Learning Decision RuleClassifications from Data Streams, IJCAI 2011, J. Gama, P. Kosina

Parameters:

• -p: Minimum value of p
• -t: Tie Threshold
• -c: Split Confidence
• -g: GracePeriod, the number of instances a leaf should observe between split attempts
• -o: Prediction function to use. Ex:FirstHit
• -r: Learn ordered or unordered rule

Version:

$Revision: 2 $

Author:

P. Kosina, E. Almeida, J. Gama

See Also:

Serialized Form

Field Summary

Fields

Modifier and Type	Field and Description
FlagOption	anomalyDetectionOption
IntOption	anomalyNumInstThresholdOption
FloatOption	anomalyProbabilityThresholdOption
protected AutoExpandVector<AttributeClassObserver>	attributeObservers
protected AutoExpandVector<AttributeClassObserver>	attributeObserversGauss
protected ArrayList<ArrayList<Integer>>	caseAnomaly
protected ArrayList<ArrayList<Integer>>	caseAnomalySupervised
IntOption	gracePeriodOption
protected weka.core.Instance	instance
protected DoubleVector	observedClassDistribution
FlagOption	orderedRulesOption
FloatOption	PminOption
MultiChoiceOption	predictionFunctionOption
FloatOption	probabilityThresholdOption
protected ArrayList<Integer>	ruleAnomaliesIndex
protected ArrayList<Integer>	ruleAnomaliesIndexSupervised
protected ArrayList<ArrayList<ArrayList<Double>>>	ruleAttribAnomalyStatistics
protected ArrayList<ArrayList<ArrayList<Double>>>	ruleAttribAnomalyStatisticsSupervised
protected DoubleVector	ruleClassIndex
protected ArrayList<RuleClassification>	ruleSet
protected ArrayList<RuleClassification>	ruleSetAnomalies
protected ArrayList<RuleClassification>	ruleSetAnomaliesSupervised
protected DoubleVector	saveBestEntropy
protected DoubleVector	saveBestEntropyNominalAttrib
protected DoubleVector	saveBestGlobalEntropy
protected ArrayList<ArrayList<Double>>	saveBestValGlobalEntropy
protected ArrayList<Double>	saveTheBest
FloatOption	splitConfidenceOption
FlagOption	Supervised
FloatOption	tieThresholdOption
FlagOption	Unsupervised

Fields inherited from class moa.classifiers.AbstractClassifier

classifierRandom, modelContext, randomSeed, randomSeedOption, trainingWeightSeenByModel

Fields inherited from class moa.options.AbstractOptionHandler

classOptionNamesToPreparedObjects, options

Constructor Summary

Constructors

Constructor and Description
RuleClassifier()

Method Summary

Methods

Modifier and Type	Method and Description
boolean	checkBestAttrib(double n, AutoExpandVector<AttributeClassObserver> observerss, DoubleVector observedClassDistribution)
double	computeAnomalySupervised(RuleClassification rl, int ruleIndex, weka.core.Instance inst)
double	computeAnomalyUnsupervised(RuleClassification rl, int ruleIndex, weka.core.Instance inst)
double	ComputeHoeffdingBound(double range, double confidence, double n)
double	computeMean(double sum, int size)
double	computeProbability(double mean, double sd, double value)
double	computeSD(double squaredVal, double val, int size)
void	createRule(weka.core.Instance inst)
double	entropy(DoubleVector ValorDistClassE)
void	expandeRule(RuleClassification rl, weka.core.Instance inst, int ruleIndex)
void	findBestValEntropy(BinaryTreeNumericAttributeClassObserver.Node node, DoubleVector classCountL, DoubleVector classCountR, boolean status, double minEntropy, DoubleVector parentCCLeft)
void	findBestValEntropyNominalAtt(AutoExpandVector<DoubleVector> attrib, int attNumValues)
protected double[]	firstHit(weka.core.Instance inst)
protected double[]	getBestSecondBestEntropy(DoubleVector entropy)
int	getCountNominalAttrib(ArrayList<Predicates> predicateSet)
void	getModelDescription(StringBuilder out, int indent) Returns a string representation of the model.
void	getModelDescriptionNoAnomalyDetection(StringBuilder out, int indent)
protected Measurement[]	getModelMeasurementsImpl() Gets the current measurements of this classifier. The reason for ...Impl methods: ease programmer burden by not requiring them to remember calls to super in overridden methods.
String	getPurposeString() Gets the purpose of this object
protected double	getRuleMajorityClassIndex(RuleClassification r)
double[]	getVotesForInstance(weka.core.Instance inst) Predicts the class memberships for a given instance.
double	getWeightSeen()
void	initializeRuleStatistics(RuleClassification rl, Predicates pred, weka.core.Instance inst)
boolean	isRandomizable() Gets whether this classifier needs a random seed.
void	mainFindBestValEntropy(BinaryTreeNumericAttributeClassObserver.Node root)
void	manageMemory(int currentByteSize, int maxByteSize)
protected AttributeClassObserver	newNominalClassObserver()
protected AttributeClassObserver	newNumericClassObserver()
protected AttributeClassObserver	newNumericClassObserver2()
protected double[]	oberversDistribProb(weka.core.Instance inst, DoubleVector classDistrib)
void	printAnomaliesSupervised(StringBuilder out, int indent)
void	printAnomaliesUnsupervised(StringBuilder out, int indent)
void	resetLearningImpl() Resets this classifier.
protected BigDecimal	round(double val)
void	theBestAttributes(weka.core.Instance instance, AutoExpandVector<AttributeClassObserver> observersParameter)
void	trainOnInstanceImpl(weka.core.Instance inst) Trains this classifier incrementally using the given instance. The reason for ...Impl methods: ease programmer burden by not requiring them to remember calls to super in overridden methods.
void	updateRuleAttribStatistics(weka.core.Instance inst, RuleClassification rl, int ruleIndex)
protected double[]	weightedMax(weka.core.Instance inst)
protected double[]	weightedSum(weka.core.Instance inst)

Methods inherited from class moa.classifiers.AbstractClassifier

contextIsCompatible, copy, correctlyClassifies, getAttributeNameString, getAWTRenderer, getClassLabelString, getClassNameString, getDescription, getModelContext, getModelMeasurements, getNominalValueString, getSubClassifiers, 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

instance

protected weka.core.Instance instance

attributeObservers

protected AutoExpandVector<AttributeClassObserver> attributeObservers

attributeObserversGauss

protected AutoExpandVector<AttributeClassObserver> attributeObserversGauss

observedClassDistribution

protected DoubleVector observedClassDistribution

saveBestEntropy

protected DoubleVector saveBestEntropy

saveBestEntropyNominalAttrib

protected DoubleVector saveBestEntropyNominalAttrib

ruleClassIndex

protected DoubleVector ruleClassIndex

saveBestGlobalEntropy

protected DoubleVector saveBestGlobalEntropy

saveBestValGlobalEntropy

protected ArrayList<ArrayList<Double>> saveBestValGlobalEntropy

saveTheBest

protected ArrayList<Double> saveTheBest

ruleSet

protected ArrayList<RuleClassification> ruleSet

ruleSetAnomalies

protected ArrayList<RuleClassification> ruleSetAnomalies

ruleAnomaliesIndex

protected ArrayList<Integer> ruleAnomaliesIndex

caseAnomaly

protected ArrayList<ArrayList<Integer>> caseAnomaly

ruleAttribAnomalyStatistics

protected ArrayList<ArrayList<ArrayList<Double>>> ruleAttribAnomalyStatistics

ruleSetAnomaliesSupervised

protected ArrayList<RuleClassification> ruleSetAnomaliesSupervised

ruleAnomaliesIndexSupervised

protected ArrayList<Integer> ruleAnomaliesIndexSupervised

caseAnomalySupervised

protected ArrayList<ArrayList<Integer>> caseAnomalySupervised

ruleAttribAnomalyStatisticsSupervised

protected ArrayList<ArrayList<ArrayList<Double>>> ruleAttribAnomalyStatisticsSupervised

PminOption

public FloatOption PminOption

splitConfidenceOption

public FloatOption splitConfidenceOption

tieThresholdOption

public FloatOption tieThresholdOption

anomalyProbabilityThresholdOption

public FloatOption anomalyProbabilityThresholdOption

probabilityThresholdOption

public FloatOption probabilityThresholdOption

anomalyNumInstThresholdOption

public IntOption anomalyNumInstThresholdOption

gracePeriodOption

public IntOption gracePeriodOption

predictionFunctionOption

public MultiChoiceOption predictionFunctionOption

orderedRulesOption

public FlagOption orderedRulesOption

anomalyDetectionOption

public FlagOption anomalyDetectionOption

Supervised

public FlagOption Supervised

Unsupervised

public FlagOption Unsupervised

Constructor Detail

RuleClassifier

public RuleClassifier()

Method Detail

getPurposeString

public String getPurposeString()

Description copied from interface: OptionHandler

Gets the purpose of this object

Specified by:

getPurposeString in interface OptionHandler

Overrides:

getPurposeString in class AbstractClassifier

Returns:

the string with the purpose of this object

getVotesForInstance

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

Description copied from interface: Classifier

Predicts the class memberships for a given instance. If an instance is unclassified, the returned array elements must be all zero.

Parameters:

inst - the instance to be classified

Returns:

an array containing the estimated membership probabilities of the test instance in each class

getModelMeasurementsImpl

protected Measurement[] getModelMeasurementsImpl()

Description copied from class: AbstractClassifier

Gets the current measurements of this classifier.

The reason for ...Impl methods: ease programmer burden by not requiring them to remember calls to super in overridden methods. Note that this will produce compiler errors if not overridden.

Specified by:

getModelMeasurementsImpl in class AbstractClassifier

Returns:

an array of measurements to be used in evaluation tasks

resetLearningImpl

public void resetLearningImpl()

Description copied from class: AbstractClassifier

Resets this classifier. It must be similar to starting a new classifier from scratch.

The reason for ...Impl methods: ease programmer burden by not requiring them to remember calls to super in overridden methods. Note that this will produce compiler errors if not overridden.

Specified by:

resetLearningImpl in class AbstractClassifier

getWeightSeen

public double getWeightSeen()

trainOnInstanceImpl

public void trainOnInstanceImpl(weka.core.Instance inst)

Description copied from class: AbstractClassifier

Trains this classifier incrementally using the given instance.

The reason for ...Impl methods: ease programmer burden by not requiring them to remember calls to super in overridden methods. Note that this will produce compiler errors if not overridden.

Specified by:

trainOnInstanceImpl in class AbstractClassifier

Parameters:

inst - the instance to be used for training

getModelDescription

public void getModelDescription(StringBuilder out,
                       int indent)

Description copied from class: AbstractClassifier

Returns a string representation of the model.

Specified by:

getModelDescription in class AbstractClassifier

Parameters:

out - the stringbuilder to add the description

indent - the number of characters to indent

printAnomaliesUnsupervised

public void printAnomaliesUnsupervised(StringBuilder out,
                              int indent)

printAnomaliesSupervised

public void printAnomaliesSupervised(StringBuilder out,
                            int indent)

getModelDescriptionNoAnomalyDetection

public void getModelDescriptionNoAnomalyDetection(StringBuilder out,
                                         int indent)

isRandomizable

public boolean isRandomizable()

Description copied from interface: Classifier

Gets whether this classifier needs a random seed. Examples of methods that needs a random seed are bagging and boosting.

Returns:

true if the classifier needs a random seed.

getCountNominalAttrib

public int getCountNominalAttrib(ArrayList<Predicates> predicateSet)

round

protected BigDecimal round(double val)

initializeRuleStatistics

public void initializeRuleStatistics(RuleClassification rl,
                            Predicates pred,
                            weka.core.Instance inst)

updateRuleAttribStatistics

public void updateRuleAttribStatistics(weka.core.Instance inst,
                              RuleClassification rl,
                              int ruleIndex)

computeAnomalyUnsupervised

public double computeAnomalyUnsupervised(RuleClassification rl,
                                int ruleIndex,
                                weka.core.Instance inst)

computeAnomalySupervised

public double computeAnomalySupervised(RuleClassification rl,
                              int ruleIndex,
                              weka.core.Instance inst)

computeMean

public double computeMean(double sum,
                 int size)

computeSD

public double computeSD(double squaredVal,
               double val,
               int size)

computeProbability

public double computeProbability(double mean,
                        double sd,
                        double value)

createRule

public void createRule(weka.core.Instance inst)

expandeRule

public void expandeRule(RuleClassification rl,
               weka.core.Instance inst,
               int ruleIndex)

theBestAttributes

public void theBestAttributes(weka.core.Instance instance,
                     AutoExpandVector<AttributeClassObserver> observersParameter)

entropy

public double entropy(DoubleVector ValorDistClassE)

findBestValEntropy

public void findBestValEntropy(BinaryTreeNumericAttributeClassObserver.Node node,
                      DoubleVector classCountL,
                      DoubleVector classCountR,
                      boolean status,
                      double minEntropy,
                      DoubleVector parentCCLeft)

mainFindBestValEntropy

public void mainFindBestValEntropy(BinaryTreeNumericAttributeClassObserver.Node root)

findBestValEntropyNominalAtt

public void findBestValEntropyNominalAtt(AutoExpandVector<DoubleVector> attrib,
                                int attNumValues)

ComputeHoeffdingBound

public double ComputeHoeffdingBound(double range,
                           double confidence,
                           double n)

checkBestAttrib

public boolean checkBestAttrib(double n,
                      AutoExpandVector<AttributeClassObserver> observerss,
                      DoubleVector observedClassDistribution)

getBestSecondBestEntropy

protected double[] getBestSecondBestEntropy(DoubleVector entropy)

getRuleMajorityClassIndex

protected double getRuleMajorityClassIndex(RuleClassification r)

oberversDistribProb

protected double[] oberversDistribProb(weka.core.Instance inst,
                           DoubleVector classDistrib)

firstHit

protected double[] firstHit(weka.core.Instance inst)

weightedMax

protected double[] weightedMax(weka.core.Instance inst)

weightedSum

protected double[] weightedSum(weka.core.Instance inst)

newNominalClassObserver

protected AttributeClassObserver newNominalClassObserver()

newNumericClassObserver

protected AttributeClassObserver newNumericClassObserver()

newNumericClassObserver2

protected AttributeClassObserver newNumericClassObserver2()

manageMemory

public void manageMemory(int currentByteSize,
                int maxByteSize)