Package net.sf.tweety.logics.pcl.analysis

Class DistanceMinimizationInconsistencyMeasure

java.lang.Object

net.sf.tweety.logics.commons.analysis.BeliefSetInconsistencyMeasure<ProbabilisticConditional>

net.sf.tweety.logics.pcl.analysis.DistanceMinimizationInconsistencyMeasure

All Implemented Interfaces:

PostulateEvaluatable<ProbabilisticConditional>, InconsistencyMeasure<BeliefSet<ProbabilisticConditional,?>>

Direct Known Subclasses:

NormalizedDistanceMinimizationInconsistencyMeasure

public class DistanceMinimizationInconsistencyMeasure
extends BeliefSetInconsistencyMeasure<ProbabilisticConditional>

This class models the distance minimization inconsistency measure as proposed in [Thimm,UAI,2009], extended by the use of different p-norms.

Author:

Matthias Thimm

Field Summary

Fields

Modifier and Type	Field	Description
private java.util.Map<PclBeliefSet,java.lang.Double>	archive	For archiving inconsistency values.
private java.util.Map<PclBeliefSet,java.util.Map<ProbabilisticConditional,java.lang.Double>>	archiveDevs	For archiving eta/tau values.
private static org.slf4j.Logger	log	Logger.
private int	p	The p-norm parameter.

Fields inherited from interface net.sf.tweety.logics.commons.analysis.InconsistencyMeasure

MEASURE_TOLERANCE

Constructor Summary

Constructors

Constructor	Description
DistanceMinimizationInconsistencyMeasure()	Creates a new measure for p=1.
DistanceMinimizationInconsistencyMeasure(int p)	Creates a new measure for the given p.

Method Summary

Modifier and Type	Method	Description
java.lang.Double	getDeviation(PclBeliefSet beliefSet, ProbabilisticConditional pc)	Returns the deviation of the given conditional in the nearest consistent belief set.
java.lang.Double	inconsistencyMeasure(java.util.Collection<ProbabilisticConditional> formulas)	This method measures the inconsistency of the given set of formulas.

Methods inherited from class net.sf.tweety.logics.commons.analysis.BeliefSetInconsistencyMeasure

inconsistencyMeasure

Methods inherited from class java.lang.Object

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

Field Detail

log

private static org.slf4j.Logger log

Logger.

p

private int p

The p-norm parameter.

archive

private java.util.Map<PclBeliefSet,java.lang.Double> archive

For archiving inconsistency values.

archiveDevs

private java.util.Map<PclBeliefSet,java.util.Map<ProbabilisticConditional,java.lang.Double>> archiveDevs

For archiving eta/tau values.

Constructor Detail

DistanceMinimizationInconsistencyMeasure

public DistanceMinimizationInconsistencyMeasure()

Creates a new measure for p=1.

DistanceMinimizationInconsistencyMeasure

public DistanceMinimizationInconsistencyMeasure(int p)

Creates a new measure for the given p.

Parameters:

p - some parameter for the p-norm.

Method Detail

getDeviation

public java.lang.Double getDeviation(PclBeliefSet beliefSet,
                                     ProbabilisticConditional pc)

Returns the deviation of the given conditional in the nearest consistent belief set.

Parameters:

beliefSet - some belief set.

pc - a probabilistic conditional.

Returns:

a double.

inconsistencyMeasure

public java.lang.Double inconsistencyMeasure(java.util.Collection<ProbabilisticConditional> formulas)

Description copied from class: BeliefSetInconsistencyMeasure

This method measures the inconsistency of the given set of formulas.

Specified by:

inconsistencyMeasure in class BeliefSetInconsistencyMeasure<ProbabilisticConditional>

Parameters:

formulas - a collection of formulas.

Returns:

a Double indicating the degree of inconsistency.