Package org.tweetyproject.arg.dung.causal.syntax

Class CausalModel

java.lang.Object

org.tweetyproject.commons.BeliefSet<PlFormula,PlSignature>

org.tweetyproject.logics.pl.syntax.PlBeliefSet

org.tweetyproject.arg.dung.causal.syntax.CausalModel

All Implemented Interfaces:

Iterable<PlFormula>, Collection<PlFormula>, BeliefBase

public class CausalModel
extends PlBeliefSet

This class describes a causal model. Reference: "Argumentation-based Causal and Counterfactual Reasoning" by Lars Bengel, Lydia Blümel, Tjitze Rienstra and Matthias Thimm, published at 1st International Workshop on Argumentation for eXplainable AI (ArgXAI, co-located with COMMA ’22), September 12, 2022

Version:

TweetyProject 1.23

Author:

Julian Sander

Field Summary

Fields inherited from class org.tweetyproject.commons.BeliefSet

EQUALS_USES_SIGNATURE

Constructor Summary

Constructors

Constructor	Description
CausalModel()	Creates an empty causal model
CausalModel(Set<Equivalence> structuralEquations)	Creates a causal model
CausalModel(Set<Proposition> backGroundAtoms, Set<Proposition> explainableAtoms, Set<Equivalence> structuralEquations)	Creates a causal model

Method Summary

Modifier and Type	Method	Description
boolean	add(PlFormula formula)
boolean	add(PlFormula... formulas)	Adds the specified elements to the end of this collection (optional operation).
boolean	addAll(Collection<? extends PlFormula> formulas)
boolean	addBackgroundAtom(Proposition atom)	Adds a specified atom to the background atoms of this instance.
boolean	addExplainableAtom(Proposition atom, Equivalence structuralEquation)	Adds a specified atom to the explainable atoms of this instance and the specified equivalence as a structural equation for this atom.
boolean	addStructuralEquation(Equivalence equation)	Adds a specified equivalence to the structural equations of this instance.
CausalModel	clone()
boolean	equals(Object o)
HashSet<Proposition>	getBackGroundAtoms()	Retrieves a copy of the set of background atoms currently defined in the causal model.
HashSet<Proposition>	getExplainableAtoms()	Retrieves a copy of the set of explainable atoms currently defined in the causal model.
HashSet<Equivalence>	getStructuralEquations()	Retrieves a copy of the set of structural equations currently defined in the causal model.
CausalModel	getTwinModel()
int	hashCode()
void	intervene(Proposition v, boolean x)	This method implements the interventional statement, by replacing all structural equations for the specified explainable atom "v" of this causal model with a formula "v = x".

Methods inherited from class org.tweetyproject.logics.pl.syntax.PlBeliefSet

getCanonicalOrdering, getMinimalSignature, getSyntaxComponents, toCnf

Methods inherited from class org.tweetyproject.commons.BeliefSet

clear, contains, containsAll, getSignature, isEmpty, iterator, remove, removeAll, retainAll, setSignature, size, toArray, toArray, toString, toString

Methods inherited from class java.lang.Object

getClass, notify, notifyAll, wait, wait, wait

Methods inherited from interface java.util.Collection

parallelStream, removeIf, spliterator, stream, toArray

Methods inherited from interface java.lang.Iterable

forEach

Constructor Details

CausalModel

public CausalModel()

Creates an empty causal model

CausalModel

public CausalModel(Set<Equivalence> structuralEquations)

Creates a causal model

Parameters:

structuralEquations - Set of boolean structural equations; One equation for each explainable atom, which only occurs on the left side of the equation

CausalModel

public CausalModel(Set<Proposition> backGroundAtoms,
 Set<Proposition> explainableAtoms,
 Set<Equivalence> structuralEquations)

Creates a causal model

Parameters:

backGroundAtoms - Set of background atoms

explainableAtoms - Set of explainable atoms

structuralEquations - Set of boolean structural equations; One equation for each explainable atom, which only occurs on the left side of the equation

Method Details

add

public boolean add(PlFormula formula)

Specified by:

add in interface Collection<PlFormula>

Overrides:

add in class BeliefSet<PlFormula,PlSignature>

add

public boolean add(PlFormula... formulas)

Description copied from class: BeliefSet

Adds the specified elements to the end of this collection (optional operation).

Overrides:

add in class BeliefSet<PlFormula,PlSignature>

Parameters:

formulas - to be appended to collection

Returns:

true if all elements were added, false otherwise

addAll

public boolean addAll(Collection<? extends PlFormula> formulas)

Specified by:

addAll in interface Collection<PlFormula>

Overrides:

addAll in class BeliefSet<PlFormula,PlSignature>

addBackgroundAtom

public boolean addBackgroundAtom(Proposition atom)

Adds a specified atom to the background atoms of this instance.

Parameters:

atom - Atom to add to the background atoms of this instance.

Returns:

TRUE iff the atom was successfully added as a background atom to this instance. FALSE if the atom was already contained and leaves the set unchanged.

Throws:

IllegalArgumentException - if adding the specified atom would violate the definition of a causal model

addExplainableAtom

public boolean addExplainableAtom(Proposition atom,
 Equivalence structuralEquation)

Adds a specified atom to the explainable atoms of this instance and the specified equivalence as a structural equation for this atom.

Parameters:

atom - Atom to add to the explainable atoms of this instance.

structuralEquation - Equation, having only the specified atom on the left side and a formula consisting only of atoms different than the specified atom on the right hand side of this equivalence

Returns:

TRUE iff the atom and the formula were successfully added to this instance. FALSE if the atom or the equation were already contained and leaves at least one of the sets unchanged.

Throws:

IllegalArgumentException - if adding the specified atom and or equation would violate the definition of a causal model

addStructuralEquation

public boolean addStructuralEquation(Equivalence equation)

Adds a specified equivalence to the structural equations of this instance. To this instance unknown explainable or background atoms are created and added in this process.

Parameters:

equation - Equation, having only one explainable atom on the left side and a formula on the right hand side of this equivalence consisting only of atoms different than the atom on the left

Returns:

TRUE iff the formula and all therein contained atoms were successfully added to this instance. FALSE if the equation was already contained and leaves the set unchanged.

Throws:

IllegalArgumentException - if adding the specified equation would violate the definition of a causal model

equals

public boolean equals(Object o)

Specified by:

equals in interface Collection<PlFormula>

Overrides:

equals in class BeliefSet<PlFormula,PlSignature>

getBackGroundAtoms

public HashSet<Proposition> getBackGroundAtoms()

Retrieves a copy of the set of background atoms currently defined in the causal model. Background atoms are those elements that are used in the model's equations but are not directly explainable. Returning a copy of the set ensures that the internal state of the causal model cannot be altered directly.

Returns:

A new HashSet containing all the background atoms.

getExplainableAtoms

public HashSet<Proposition> getExplainableAtoms()

Retrieves a copy of the set of explainable atoms currently defined in the causal model. Explainable atoms are those elements that the model explains directly through its structural equations. Returning a copy of the set helps maintain encapsulation and prevents external modifications to the model's state.

Returns:

A new HashSet containing all the explainable atoms.

getStructuralEquations

public HashSet<Equivalence> getStructuralEquations()

Retrieves a copy of the set of structural equations currently defined in the causal model. Structural equations are representations of equivalences that define how certain propositions (explainable atoms) depend on other atoms (background or explainable). Each equation must be an instance of Equivalence. This method filters the model's formulas to return only those that are instances of Equivalence.

Returns:

A new HashSet containing all the structural equations.

getTwinModel

public CausalModel getTwinModel()

Returns:

A twin model to this instance.

hashCode

public int hashCode()

Specified by:

hashCode in interface Collection<PlFormula>

Overrides:

hashCode in class BeliefSet<PlFormula,PlSignature>

intervene

public void intervene(Proposition v,
 boolean x)

This method implements the interventional statement, by replacing all structural equations for the specified explainable atom "v" of this causal model with a formula "v = x".

Parameters:

v - Explainable atom of this causal model, which is to be intervened on.

x - Truth value of the intervention.

clone

public CausalModel clone()