edu.udo.cs.yale.operator.learner.meta
Class SDRulesetInduction

java.lang.Object
  edu.udo.cs.yale.operator.Operator
      edu.udo.cs.yale.operator.OperatorChain
          edu.udo.cs.yale.operator.learner.meta.SDRulesetInduction

All Implemented Interfaces:

ConfigurationListener

public class SDRulesetInduction
extends OperatorChain

Subgroup discovery learner.

Version:

$Id: SDRulesetInduction.java,v 1.20 2006/04/05 08:57:26 ingomierswa Exp $

Author:

Martin Scholz

Field Summary

private int	currentIteration
static java.lang.String	GAMMA Boolean parameter to specify whether the label priors should be equally likely after first iteration.
static java.lang.String	INTERNAL_BOOTSTRAP Name of the flag indicating internal bootstrapping.
static double	MIN_ADVANTAGE Discard models with an advantage of less than the specified value.
static java.lang.String	NUM_OF_ITERATIONS Name of the variable specifying the maximal number of iterations of the learner.
private double	performance
static java.lang.String	REWEIGHT Boolean parameter: true for additive reweighting, false for multiplicative.
static java.lang.String	ROC_FILTER A parameter whether to discard all rules not lying on the convex hull in ROC space.
static java.lang.String	TIMES_COVERED Name of special attribute counting the times an example has been covered by a rule.

Constructor Summary

SDRulesetInduction(OperatorDescription description)
Constructor.

Method Summary

IOObject[]	apply() Constructs a Model repeatedly running a weak learner, reweighting the training example set accordingly, and combining the hypothesis using the available weighted performance values.
private void	debugMessage(SDReweightMeasures wp)
InnerOperatorCondition	getInnerOperatorCondition() Must return a condition of the IO behaviour of all desired inner operators.
java.lang.Class[]	getInputClasses() Returns the classes that are needed as input.
int	getMaxNumberOfInnerOperators() Returns the maximum number of innner operators.
int	getMinNumberOfInnerOperators() Returns the minimum number of innner operators.
int	getNumberOfSteps() Returns the number of steps performed by this chain.
java.lang.Class[]	getOutputClasses() Returns the classes that are guaranteed to be returned by apply() as additional output.
java.util.List<ParameterType>	getParameterTypes() Adds the parameters "number of iterations" and "model file".
static int	getPosIndex(Attribute label)
private boolean	isOnConvexHull(java.util.List<double[]> rocCurve, double tpr, double fpr)
private double[]	prepareWeights(ExampleSet exampleSet) Creates a weight attribute if not yet done and fills it with an initial value so that positive and negative examples are equally probable.
private PredictionModel	trainModel(ExampleSet exampleSet) Runs the "embedded" learner on the example set and retuns a model.
private SDEnsemble	trainRuleset(ExampleSet trainingSet, double[] classPriors) Main method for training the ensemble classifier

Methods inherited from class edu.udo.cs.yale.operator.OperatorChain

addAddListener, addOperator, addOperator, checkDeprecations, checkIO, checkNumberOfInnerOperators, checkProperties, clearErrorList, clearStepCounter, cloneOperator, countStep, createExperimentTree, delete, experimentFinished, experimentStarts, getAllInnerOperators, getCurrentStep, getIndexOfOperator, getInnerOperatorForName, getInnerOperatorsXML, getNumberOfAllOperators, getNumberOfChildrensSteps, getNumberOfOperators, getOperator, getOperatorFromAll, getOperators, isEnabled, performAdditionalChecks, removeAddListener, removeOperator, setEnabled, setExperiment, shouldReturnInnerOutput

Methods inherited from class edu.udo.cs.yale.operator.Operator

addError, addValue, addWarning, apply, createExperimentTree, createFromXML, createMarkedExperimentTree, getAddOnlyAdditionalOutput, getApplyCount, getDeliveredOutputClasses, getDeprecationInfo, getDesiredInputClasses, getErrorList, getExperiment, getInput, getInput, getInput, getInputDescription, getIOContainerForInApplyLoopBreakpoint, getName, getOperatorClassName, getOperatorDescription, getParameter, getParameterAsBoolean, getParameterAsColor, getParameterAsDouble, getParameterAsFile, getParameterAsInt, getParameterAsString, getParameterList, getParameters, getParameterType, getParent, getStartTime, getStatus, getUserDescription, getValue, getValues, getXML, hasBreakpoint, hasBreakpoint, hasInput, inApplyLoop, isParameterSet, logMessage, register, remove, rename, resume, setBreakpoint, setInput, setListParameter, setOperatorParameters, setParameter, setParameters, setParent, setUserDescription, toString, writeXML

Methods inherited from class java.lang.Object

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

Field Detail

NUM_OF_ITERATIONS

public static final java.lang.String NUM_OF_ITERATIONS

Name of the variable specifying the maximal number of iterations of the learner.

See Also:

Constant Field Values

INTERNAL_BOOTSTRAP

public static final java.lang.String INTERNAL_BOOTSTRAP

Name of the flag indicating internal bootstrapping.

See Also:

Constant Field Values

ROC_FILTER

public static final java.lang.String ROC_FILTER

A parameter whether to discard all rules not lying on the convex hull in ROC space.

See Also:

Constant Field Values

REWEIGHT

public static final java.lang.String REWEIGHT

Boolean parameter: true for additive reweighting, false for multiplicative.

See Also:

Constant Field Values

GAMMA

public static final java.lang.String GAMMA

Boolean parameter to specify whether the label priors should be equally likely after first iteration.

See Also:

Constant Field Values

TIMES_COVERED

public static final java.lang.String TIMES_COVERED

Name of special attribute counting the times an example has been covered by a rule. This attribute is created for additive reweighting, only.

See Also:

Constant Field Values

MIN_ADVANTAGE

public static final double MIN_ADVANTAGE

Discard models with an advantage of less than the specified value.

See Also:

Constant Field Values

performance

private double performance

currentIteration

private int currentIteration

Constructor Detail

SDRulesetInduction

public SDRulesetInduction(OperatorDescription description)

Constructor.

Method Detail

getInnerOperatorCondition

public InnerOperatorCondition getInnerOperatorCondition()

Description copied from class: OperatorChain

Must return a condition of the IO behaviour of all desired inner operators. If there are no "special" conditions and the chain works similar to a simple operator chain this method should at least return a SimpleChainInnerOperatorCondition. More than one condition should be combined with help of the class CombinedInnerOperatorCondition.

Specified by:

getInnerOperatorCondition in class OperatorChain

getMaxNumberOfInnerOperators

public int getMaxNumberOfInnerOperators()

Description copied from class: OperatorChain

Returns the maximum number of innner operators.

Specified by:

getMaxNumberOfInnerOperators in class OperatorChain

See Also:

OperatorChain.getMaxNumberOfInnerOperators()

getMinNumberOfInnerOperators

public int getMinNumberOfInnerOperators()

Description copied from class: OperatorChain

Returns the minimum number of innner operators.

Specified by:

getMinNumberOfInnerOperators in class OperatorChain

See Also:

OperatorChain.getMinNumberOfInnerOperators()

getNumberOfSteps

public int getNumberOfSteps()

Description copied from class: OperatorChain

Returns the number of steps performed by this chain.

Specified by:

getNumberOfSteps in class OperatorChain

See Also:

OperatorChain.getNumberOfSteps()

getInputClasses

public java.lang.Class[] getInputClasses()

Description copied from class: Operator

Returns the classes that are needed as input. May be null or an empty (no desired input). As default, all delivered input objects are consumed and must be also delivered as output in both Operator.getOutputClasses() and Operator.apply() if this is necessary. This default behavior can be changed by overriding Operator.getInputDescription(Class). Subclasses which implement this method should not make use of parameters since this method is invoked by getParameterTypes(). Therefore, parameters are not fully available at this point of time and this might lead to exceptions. Please use InputDescriptions instead.

Specified by:

getInputClasses in class Operator

See Also:

Operator.getInputClasses()

getOutputClasses

public java.lang.Class[] getOutputClasses()

Description copied from class: Operator

Returns the classes that are guaranteed to be returned by apply() as additional output. Please note that input object which should not be consumed must also be defined by this method (e.g. for preprocessing operators). The default behavior for input consumation is defined by Operator.getInputDescription(Class) and can be changed by overwriting this method. Objects which are not consumed must not be defined as additional output in this method. May be null or an empy array (no additional output is produced).

Specified by:

getOutputClasses in class Operator

See Also:

Operator.getOutputClasses()

getPosIndex

public static int getPosIndex(Attribute label)

prepareWeights

private double[] prepareWeights(ExampleSet exampleSet)
                         throws OperatorException

Creates a weight attribute if not yet done and fills it with an initial value so that positive and negative examples are equally probable.

Parameters:

exampleSet - the example set to be prepared

Throws:

OperatorException

trainModel

private PredictionModel trainModel(ExampleSet exampleSet)
                            throws OperatorException

Runs the "embedded" learner on the example set and retuns a model.

Parameters:

exampleSet - an ExampleSet to train a model for

Returns:

a Model

Throws:

OperatorException

apply

public IOObject[] apply()
                 throws OperatorException

Constructs a Model repeatedly running a weak learner, reweighting the training example set accordingly, and combining the hypothesis using the available weighted performance values. If the input contains a model, then this model is used as a starting point for weighting the examples.

Overrides:

apply in class OperatorChain

Returns:

the last inner operator's output or the input itself if the chain is empty.

Throws:

OperatorException

trainRuleset

private SDEnsemble trainRuleset(ExampleSet trainingSet,
                                double[] classPriors)
                         throws OperatorException

Main method for training the ensemble classifier

Throws:

OperatorException

debugMessage

private void debugMessage(SDReweightMeasures wp)

isOnConvexHull

private boolean isOnConvexHull(java.util.List<double[]> rocCurve,
                               double tpr,
                               double fpr)

getParameterTypes

public java.util.List<ParameterType> getParameterTypes()

Adds the parameters "number of iterations" and "model file".

Overrides:

getParameterTypes in class Operator