Package net.sf.tweety.logics.pcl.reasoner

Class GeneralizedMeReasoner

java.lang.Object

net.sf.tweety.logics.pcl.reasoner.AbstractPclReasoner

net.sf.tweety.logics.pcl.reasoner.GeneralizedMeReasoner

All Implemented Interfaces:

ModelProvider<ProbabilisticConditional,PclBeliefSet,ProbabilityDistribution<PossibleWorld>>, QuantitativeReasoner<PclBeliefSet,PlFormula>, Reasoner<java.lang.Double,PclBeliefSet,PlFormula>

public class GeneralizedMeReasoner
extends AbstractPclReasoner

This class implements a generalized maximum entropy reasoner for probabilistic conditional logic as proposed in [Potyka, Thimm, 2014] which also works for inconsistent belief sets. It computes the generalized ME-distribution for the given belief set and answers queries with respect to this ME-distribution.

Author:

Matthias Thimm

Field Summary

Fields

Modifier and Type	Field	Description
private double	accuracy	The numerical accuracy.
static int	EUCLIDEAN
private BeliefSetInconsistencyMeasure<ProbabilisticConditional>	inc	The corresponding inconsistency measure.
static int	MANHATTAN
static int	MAXIMUM
private RealVectorNorm	norm	The norm.

Constructor Summary

Constructors

Constructor	Description
GeneralizedMeReasoner(int p)	Creates a new generalized ME-reasoner

Method Summary

Modifier and Type	Method	Description
ProbabilityDistribution<PossibleWorld>	getModel(PclBeliefSet beliefbase)	Returns a single (dedicated) model of the given belief base.
ProbabilityDistribution<PossibleWorld>	getModel(PclBeliefSet bs, PlSignature signature)	Computes the ME-distribution this reasoner bases on.
java.util.Collection<ProbabilityDistribution<PossibleWorld>>	getModels(PclBeliefSet bbase)	Returns a characterizing model of the given belief base
java.lang.Double	query(PclBeliefSet beliefbase, PlFormula formula)	Queries the given belief base for the given formula.

Methods inherited from class java.lang.Object

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

Field Detail

MANHATTAN

public static final int MANHATTAN

See Also:

Constant Field Values

EUCLIDEAN

public static final int EUCLIDEAN

See Also:

Constant Field Values

MAXIMUM

public static final int MAXIMUM

See Also:

Constant Field Values

norm

private RealVectorNorm norm

The norm.

inc

private BeliefSetInconsistencyMeasure<ProbabilisticConditional> inc

The corresponding inconsistency measure.

accuracy

private double accuracy

The numerical accuracy.

Constructor Detail

GeneralizedMeReasoner

public GeneralizedMeReasoner(int p)

Creates a new generalized ME-reasoner

Parameters:

p - the p-norm used

Method Detail

query

public java.lang.Double query(PclBeliefSet beliefbase,
                              PlFormula formula)

Description copied from interface: Reasoner

Queries the given belief base for the given formula.

Specified by:

query in interface Reasoner<java.lang.Double,PclBeliefSet,PlFormula>

Specified by:

query in class AbstractPclReasoner

Parameters:

beliefbase - a belief base

formula - a formula

Returns:

the answer to the query

getModels

public java.util.Collection<ProbabilityDistribution<PossibleWorld>> getModels(PclBeliefSet bbase)

Description copied from interface: ModelProvider

Returns a characterizing model of the given belief base

Specified by:

getModels in interface ModelProvider<ProbabilisticConditional,PclBeliefSet,ProbabilityDistribution<PossibleWorld>>

Specified by:

getModels in class AbstractPclReasoner

Parameters:

bbase - some belief base

Returns:

the (selected) models of the belief base

getModel

public ProbabilityDistribution<PossibleWorld> getModel(PclBeliefSet beliefbase)

Description copied from interface: ModelProvider

Returns a single (dedicated) model of the given belief base. If the implemented method allows for more than one dedicated model, the selection may be non-deterministic.

Specified by:

getModel in interface ModelProvider<ProbabilisticConditional,PclBeliefSet,ProbabilityDistribution<PossibleWorld>>

Specified by:

getModel in class AbstractPclReasoner

Parameters:

beliefbase - some belief base

Returns:

a selected model of the belief base.

getModel

public ProbabilityDistribution<PossibleWorld> getModel(PclBeliefSet bs,
                                                       PlSignature signature)

Computes the ME-distribution this reasoner bases on.

Parameters:

bs - the belief set

signature - the signature

Returns:

the ME-distribution this reasoner bases on.