net.sf.tweety.arg.dung

Class GrossiModgilRankingReasoner

java.lang.Object

net.sf.tweety.commons.Reasoner

net.sf.tweety.arg.dung.GrossiModgilRankingReasoner

public class GrossiModgilRankingReasoner
extends Reasoner

This class implements the argument ranking approach of [Grossi, Modgil. On the Graded Acceptability of Arguments. IJCAI 2015] The used algorithms are simple brute force search algorithms.

Author:

Matthias Thimm

Constructor Summary

Constructors

Constructor and Description
GrossiModgilRankingReasoner(DungTheory theory) Creates a new reasoner for the given Dung theory

Method Summary

Modifier and Type	Method and Description
java.util.Collection<Extension>	getAllMNCompleteExtensions(int m, int n) Returns all mn-complete extensions.
java.util.Collection<Extension>	getAllMNGroundedExtensions(int m, int n) Returns all mn-grounded extensions.
java.util.Collection<Extension>	getAllMNPreferredExtensions(int m, int n) Returns all mn-preferred extensions.
java.util.Collection<Extension>	getAllMStableExtensions(int m) Returns all m-stable extensions.
ArgumentRanking	getCompleteRanking() Returns the ranking wrt.
java.util.Collection<Argument>	gradedDefense(java.util.Collection<Argument> args, int m, int n) Implements the graded defense function from Def.
java.util.Collection<Argument>	gradedNeutrality(java.util.Collection<Argument> args, int m) Implements the graded neutrality function from Def.
private boolean	isContainedInAll(Argument y, java.util.Collection<Extension> exts) Checks whether the given argument "y" is contained in all sets of arguments in "args".
boolean	isMConflictFree(java.util.Collection<Argument> args, int m) Checks whether the given set of arguments is m-conflict-free, cf.
boolean	isMNAdmissible(java.util.Collection<Argument> args, int m, int n) Checks whether the given set of arguments is mn-admissible, cf.
boolean	isMNComplete(java.util.Collection<Argument> args, int m, int n) Checks whether the given set of arguments is mn-complete, cf.
boolean	isMNGrounded(java.util.Collection<Argument> args, int m, int n) Checks whether the given set of arguments is mn-grounded, cf.
boolean	isMNPreferred(java.util.Collection<Argument> args, int m, int n) Checks whether the given set of arguments is mn-preferred, cf.
boolean	isMStable(java.util.Collection<Argument> args, int m) Checks whether the given set of arguments is m-stable, cf.
private int	numOfAttackers(Argument y, java.util.Collection<Argument> x) Determines the number of attackers from x to y.
Answer	query(Formula query) This method determines the answer of the given query wrt.

Methods inherited from class net.sf.tweety.commons.Reasoner

getKnowledgeBase

Methods inherited from class java.lang.Object

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

Constructor Detail

GrossiModgilRankingReasoner

public GrossiModgilRankingReasoner(DungTheory theory)

Creates a new reasoner for the given Dung theory

Parameters:

theory -

Method Detail

numOfAttackers

private int numOfAttackers(Argument y,
                           java.util.Collection<Argument> x)

Determines the number of attackers from x to y.

Parameters:

y - some argument

x - some set of arguments

Returns:

the number of attackers from x to y.

gradedDefense

public java.util.Collection<Argument> gradedDefense(java.util.Collection<Argument> args,
                                                    int m,
                                                    int n)

Implements the graded defense function from Def. 5, i.e. gradedDefense_m_n(X) = { x | there are not more than m different y s.t. y attacks x and there are not more than n different z s.t. z attacks y and z is in X}

Parameters:

args - some set of arguments

m - some integer (indicating the number of attackers)

n - some integer (indicating the number of attackers of attackers)

Returns:

the set of arguments mn-defended by the given set of arguments

gradedNeutrality

public java.util.Collection<Argument> gradedNeutrality(java.util.Collection<Argument> args,
                                                       int m)

Implements the graded neutrality function from Def. 6, i.e. gradedNeutrality_m(X) = {x | there are less than m arguments y from X that attack x}

Parameters:

args - some set of arguments

m - some integer (the number of attackers)

Returns:

the set of arguments m-neutral to args.

isMConflictFree

public boolean isMConflictFree(java.util.Collection<Argument> args,
                               int m)

Checks whether the given set of arguments is m-conflict-free, cf. Def. 9.

Parameters:

args - some set of arguments

m - some integer

Returns:

"true" iff args is m-conflict-free

isMNAdmissible

public boolean isMNAdmissible(java.util.Collection<Argument> args,
                              int m,
                              int n)

Checks whether the given set of arguments is mn-admissible, cf. Def. 9.

Parameters:

args - some set of arguments

m - some integer

n - some integer

Returns:

"true" iff args is mn-admissible

isMNComplete

public boolean isMNComplete(java.util.Collection<Argument> args,
                            int m,
                            int n)

Checks whether the given set of arguments is mn-complete, cf. Def. 9.

Parameters:

args - some set of arguments

m - some integer

n - some integer

Returns:

"true" iff args is mn-complete

isMStable

public boolean isMStable(java.util.Collection<Argument> args,
                         int m)

Checks whether the given set of arguments is m-stable, cf. Def. 9.

Parameters:

args - some set of arguments

m - some integer

Returns:

"true" iff args is m-stable

isMNGrounded

public boolean isMNGrounded(java.util.Collection<Argument> args,
                            int m,
                            int n)

Checks whether the given set of arguments is mn-grounded, cf. Def. 9.

Parameters:

args - some set of arguments

m - some integer

n - some integer

Returns:

"true" iff args is mn-grounded

isMNPreferred

public boolean isMNPreferred(java.util.Collection<Argument> args,
                             int m,
                             int n)

Checks whether the given set of arguments is mn-preferred, cf. Def. 9.

Parameters:

args - some set of arguments

m - some integer

n - some integer

Returns:

"true" iff args is mn-preferred

getAllMNCompleteExtensions

public java.util.Collection<Extension> getAllMNCompleteExtensions(int m,
                                                                  int n)

Returns all mn-complete extensions.

Parameters:

m - some integer

n - some integer

Returns:

all mn-complete extensions.

getAllMNPreferredExtensions

public java.util.Collection<Extension> getAllMNPreferredExtensions(int m,
                                                                   int n)

Returns all mn-preferred extensions.

Parameters:

m - some integer

n - some integer

Returns:

all mn-preferred extensions.

getAllMNGroundedExtensions

public java.util.Collection<Extension> getAllMNGroundedExtensions(int m,
                                                                  int n)

Returns all mn-grounded extensions.

Parameters:

m - some integer

n - some integer

Returns:

all mn-grounded extensions.

getAllMStableExtensions

public java.util.Collection<Extension> getAllMStableExtensions(int m)

Returns all m-stable extensions.

Parameters:

m - some integer

n - some integer

Returns:

all m-stable extensions.

isContainedInAll

private boolean isContainedInAll(Argument y,
                                 java.util.Collection<Extension> exts)

Checks whether the given argument "y" is contained in all sets of arguments in "args".

Parameters:

y - some argument

exts - some set of sets of arguments

Returns:

"true" iff the given argument "y" is contained in all sets of arguments in "args".

getCompleteRanking

public ArgumentRanking getCompleteRanking()

Returns the ranking wrt. complete semantics, cf. Def. 10

Returns:

the ranking wrt. complete semantics, cf. Def. 10

query

public Answer query(Formula query)

Description copied from class: Reasoner

This method determines the answer of the given query wrt. to the given knowledge base.

Specified by:

query in class Reasoner

Parameters:

query - a query.

Returns:

the answer to the query.