net.sf.tweety.logics.mln.syntax

Class MlnFormula

java.lang.Object

net.sf.tweety.logics.commons.syntax.RelationalFormula

net.sf.tweety.logics.mln.syntax.MlnFormula

All Implemented Interfaces:

Formula, ClassicalFormula, ComplexLogicalFormula, Conjuctable, Disjunctable, Invertable, LogicStructure, ProbabilityAware, QuantifiedFormula, SimpleLogicalFormula

public class MlnFormula
extends RelationalFormula

Instances of this class represent first-order formulas with a weight.

Author:

Matthias Thimm, Tim Janus

Field Summary

Fields

Modifier and Type	Field and Description
private FolFormula	formula the first-order formula.
private java.lang.Double	weight The weight of the formula (null means that the formula is strict).

Constructor Summary

Constructors

Constructor and Description
MlnFormula(FolFormula formula) Creates a new strict MLN formula with the given formula.
MlnFormula(FolFormula formula, java.lang.Double weight) Creates a new MLN formula with the given formula and weight.
MlnFormula(FolFormula formula, Probability p) Creates a new MLN formula and estimates its weight w by the given probability p using the formula w = log(p/(1-p)*f) where "f" is the ratio of the number of worlds not satisfying the formula and the worlds satisfying the formula.

Method Summary

Modifier and Type	Method and Description
RelationalFormula	clone() Creates a deep copy of this formula
Conjunction	combineWithAnd(Conjuctable f) Returns a conjunction of this and the given formula.
Disjunction	combineWithOr(Disjunctable f)
RelationalFormula	complement()
boolean	containsQuantifier() Checks whether this formula contains any quantification.
java.util.Set<FOLAtom>	getAtoms() Processes the set of all atoms which appear in this formula
FolFormula	getFormula() Returns the inner formula.
java.util.Set<Functor>	getFunctors()
java.util.Set<? extends Predicate>	getPredicates() Processes the set of all predicates which appear in this formula
java.util.Set<Variable>	getQuantifierVariables()
FolSignature	getSignature() Returns the signature of the language of this formula.
java.util.Set<Term<?>>	getTerms()
<C extends Term<?>> java.util.Set<C>	getTerms(java.lang.Class<C> cls) Processes the set containing all terms of type C.
java.util.Set<Variable>	getUnboundVariables()
Probability	getUniformProbability()
java.lang.Double	getWeight() Returns the weight.
boolean	isClosed() Checks whether this formula is closed, i.e.
boolean	isClosed(java.util.Set<Variable> boundVariables) Checks whether this formula is closed, i.e.
boolean	isLiteral()
boolean	isStrict() Returns "true" iff this formula is strict.
boolean	isWellBound() Checks whether this formula is well-bound, i.e.
boolean	isWellBound(java.util.Set<Variable> boundVariables) Checks whether this formula is well-bound, i.e.
RelationalFormula	substitute(Term<?> v, Term<?> t) Substitutes all occurrences of term "v" in this formula by term "t" and returns the new formula.
java.lang.String	toString()

Methods inherited from class net.sf.tweety.logics.commons.syntax.RelationalFormula

allGroundInstances, allSubstitutions, containsTermsOfType, exchange, getPredicateCls, getSatisfactionRatio, isGround, isWellFormed, substitute

Methods inherited from class java.lang.Object

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

Methods inherited from interface net.sf.tweety.logics.commons.syntax.interfaces.SimpleLogicalFormula

equals, hashCode

Field Detail

formula

private FolFormula formula

the first-order formula.

weight

private java.lang.Double weight

The weight of the formula (null means that the formula is strict).

Constructor Detail

MlnFormula

public MlnFormula(FolFormula formula)

Creates a new strict MLN formula with the given formula.

Parameters:

formula - the first-order formula.

MlnFormula

public MlnFormula(FolFormula formula,
                  java.lang.Double weight)

Creates a new MLN formula with the given formula and weight.

Parameters:

formula - the first-order formula.

weight - the weight of the formula (null means that the formula is strict).

MlnFormula

public MlnFormula(FolFormula formula,
                  Probability p)

Creates a new MLN formula and estimates its weight w by the given probability p using the formula w = log(p/(1-p)*f) where "f" is the ratio of the number of worlds not satisfying the formula and the worlds satisfying the formula.

Parameters:

formula - the first-order formula.

p - the intended probability of the formula.

Method Detail

combineWithAnd

public Conjunction combineWithAnd(Conjuctable f)

Description copied from interface: Conjuctable

Returns a conjunction of this and the given formula.

Specified by:

combineWithAnd in interface Conjuctable

Specified by:

combineWithAnd in class RelationalFormula

Parameters:

f - a formula to be combined with AND and this.

Returns:

a conjunction of this and the given formula.

combineWithOr

public Disjunction combineWithOr(Disjunctable f)

Specified by:

combineWithOr in interface Disjunctable

Specified by:

combineWithOr in class RelationalFormula

Parameters:

f - a formula to be combined with OR and this.

Returns:

a disjunction of this and the given formula.

complement

public RelationalFormula complement()

Specified by:

complement in interface Invertable

Specified by:

complement in class RelationalFormula

Returns:

the complement of this formula.

getPredicates

public java.util.Set<? extends Predicate> getPredicates()

Description copied from interface: SimpleLogicalFormula

Processes the set of all predicates which appear in this formula

Returns:

all predicates that appear in this formula

getAtoms

public java.util.Set<FOLAtom> getAtoms()

Description copied from interface: SimpleLogicalFormula

Processes the set of all atoms which appear in this formula

Specified by:

getAtoms in interface SimpleLogicalFormula

Specified by:

getAtoms in class RelationalFormula

Returns:

all atoms that appear in this formula.

containsQuantifier

public boolean containsQuantifier()

Description copied from interface: QuantifiedFormula

Checks whether this formula contains any quantification.

Returns:

"true" if this formula contains a quantification.

substitute

public RelationalFormula substitute(Term<?> v,
                                    Term<?> t)
                             throws java.lang.IllegalArgumentException

Description copied from class: RelationalFormula

Substitutes all occurrences of term "v" in this formula by term "t" and returns the new formula. NOTE: if "v" is a variable and bound to a quantifier then "v" is not substituted in that quantifiers inner formula.

Specified by:

substitute in interface ComplexLogicalFormula

Specified by:

substitute in class RelationalFormula

Parameters:

v - the term to be substituted.

t - the term to substitute.

Returns:

a formula where every occurrence of "v" is replaced by "t".

Throws:

java.lang.IllegalArgumentException - if "v" and "t" are of different sorts (NOTE: this exception is only thrown when "v" actually appears in this formula).

getUnboundVariables

public java.util.Set<Variable> getUnboundVariables()

Returns:

a set of of unbound variables

isClosed

public boolean isClosed()

Description copied from interface: QuantifiedFormula

Checks whether this formula is closed, i.e. whether every variables occurring in the formula is bound by a quantifier.

Returns:

"true" if this formula is closed, "false" otherwise.

isClosed

public boolean isClosed(java.util.Set<Variable> boundVariables)

Description copied from interface: QuantifiedFormula

Checks whether this formula is closed, i.e. whether every variables occurring in the formula is bound by a quantifier. Every variable in "boundVariables" is already assumed to be bound.

Parameters:

boundVariables - the variables assumed to be bound.

Returns:

"true" if this formula is closed wrt. "boundVariables", "false" otherwise.

isWellBound

public boolean isWellBound()

Description copied from interface: QuantifiedFormula

Checks whether this formula is well-bound, i.e. whether no variable bound by a quantifier is again bound by another quantifier within the first quantifier's range.

Returns:

"true" if this formula is well-bound, "false" otherwise.

isWellBound

public boolean isWellBound(java.util.Set<Variable> boundVariables)

Description copied from interface: QuantifiedFormula

Checks whether this formula is well-bound, i.e. whether no variable bound by a quantifier is again bound by another quantifier within the first quantifier range. Every variable in "boundVariables" is assumed to be bound already.

Parameters:

boundVariables - the variables assumed to be bound.

Returns:

"true" if this formula is well-bound, "false" otherwise.

toString

public java.lang.String toString()

Specified by:

toString in class RelationalFormula

getFunctors

public java.util.Set<Functor> getFunctors()

Specified by:

getFunctors in class RelationalFormula

Returns:

all functors that appear in this formula.

getFormula

public FolFormula getFormula()

Returns the inner formula.

Specified by:

getFormula in interface QuantifiedFormula

Overrides:

getFormula in class RelationalFormula

Returns:

the inner formula.

getWeight

public java.lang.Double getWeight()

Returns the weight.

Returns:

the weight.

isStrict

public boolean isStrict()

Returns "true" iff this formula is strict.

Returns:

"true" iff this formula is strict.

getUniformProbability

public Probability getUniformProbability()

Specified by:

getUniformProbability in interface ProbabilityAware

Specified by:

getUniformProbability in class RelationalFormula

Returns:

this formula's probability in the uniform distribution.

isLiteral

public boolean isLiteral()

Returns:

true if the formula represents a literal in the language or false otherwise

getTerms

public java.util.Set<Term<?>> getTerms()

Returns:

a set containing all terms of this logical structure

getTerms

public <C extends Term<?>> java.util.Set<C> getTerms(java.lang.Class<C> cls)

Description copied from interface: LogicStructure

Processes the set containing all terms of type C. This method uses the equals method of the given Class and therefore does not add terms which are sub classes of type C to the set.

Parameters:

cls - The Class structure containing type information about the searched term

Returns:

A set containing all terms of type C of this logical structure

getQuantifierVariables

public java.util.Set<Variable> getQuantifierVariables()

Returns:

a set containing all quantified variables

clone

public RelationalFormula clone()

Description copied from interface: SimpleLogicalFormula

Creates a deep copy of this formula

Specified by:

clone in interface ComplexLogicalFormula

Specified by:

clone in interface SimpleLogicalFormula

Specified by:

clone in class RelationalFormula

getSignature

public FolSignature getSignature()

Description copied from interface: Formula

Returns the signature of the language of this formula.

Specified by:

getSignature in interface Formula

Specified by:

getSignature in class RelationalFormula

Returns:

the signature of the language of this formula.