aima.core.probability.bayes

Interface ConditionalProbabilityDistribution

All Known Subinterfaces:

ConditionalProbabilityTable

All Known Implementing Classes:

CPT

public interface ConditionalProbabilityDistribution

A conditional probability distribution on a RandomVariable X_i:

P(X_i | Parents(X_i)) that quantifies the effect of the parents on X_i.

Author:

Ciaran O'Reilly

Method Summary

Modifier and Type	Method and Description
boolean	contains(RandomVariable rv)
ProbabilityDistribution	getConditioningCase(AssignmentProposition... parentValues) A conditioning case is just a possible combination of values for the parent nodes - a miniature possible world, if you like.
ProbabilityDistribution	getConditioningCase(java.lang.Object... parentValues) A conditioning case is just a possible combination of values for the parent nodes - a miniature possible world, if you like.
java.util.Set<RandomVariable>	getFor() A convenience method for merging the elements of getParents() and getOn() into a consistent ordered set (i.e.
RandomVariable	getOn()
java.util.Set<RandomVariable>	getParents()
java.lang.Object	getSample(double probabilityChoice, AssignmentProposition... parentValues) Retrieve a specific example for the Random Variable this conditional distribution is on.
java.lang.Object	getSample(double probabilityChoice, java.lang.Object... parentValues) Retrieve a specific example for the Random Variable this conditional distribution is on.
double	getValue(AssignmentProposition... eventValues) Get the value for the provided set of AssignmentPropositions for the random variables comprising the Conditional Distribution (size of each must equal and their random variables must match).
double	getValue(java.lang.Object... eventValues) Get the value for the provided set of values for the random variables comprising the Conditional Distribution (ordering and size of each must equal getFor() and their domains must match).

Method Detail

getOn

RandomVariable getOn()

Returns:

the Random Variable this conditional probability distribution is on.

getParents

java.util.Set<RandomVariable> getParents()

Returns:

a consistent ordered Set (e.g. LinkedHashSet) of the parent Random Variables for this conditional distribution.

getFor

java.util.Set<RandomVariable> getFor()

A convenience method for merging the elements of getParents() and getOn() into a consistent ordered set (i.e. getOn() should always be the last Random Variable returned when iterating over the set).

Returns:

a consistent ordered Set (e.g. LinkedHashSet) of the random variables this conditional probability distribution is for.

contains

boolean contains(RandomVariable rv)

Parameters:

rv - the Random Variable to be checked.

Returns:

true if the conditional distribution is for the passed in Random Variable, false otherwise.

getValue

double getValue(java.lang.Object... eventValues)

Get the value for the provided set of values for the random variables comprising the Conditional Distribution (ordering and size of each must equal getFor() and their domains must match).

Parameters:

eventValues - the values for the random variables comprising the Conditional Distribution

Returns:

the value for the possible worlds associated with the assignments for the random variables comprising the Conditional Distribution.

getValue

double getValue(AssignmentProposition... eventValues)

Get the value for the provided set of AssignmentPropositions for the random variables comprising the Conditional Distribution (size of each must equal and their random variables must match).

Parameters:

eventValues - the assignment propositions for the random variables comprising the Conditional Distribution

Returns:

the value for the possible worlds associated with the assignments for the random variables comprising the Conditional Distribution.

getConditioningCase

ProbabilityDistribution getConditioningCase(java.lang.Object... parentValues)

A conditioning case is just a possible combination of values for the parent nodes - a miniature possible world, if you like. The returned distribution must sum to 1, because the entries represent an exhaustive set of cases for the random variable.

Parameters:

parentValues - for the conditioning case. The ordering and size of parentValues must equal getParents() and their domains must match.

Returns:

the Probability Distribution for the Random Variable the Conditional Probability Distribution is On.

See Also:

getOn(), getParents()

getConditioningCase

ProbabilityDistribution getConditioningCase(AssignmentProposition... parentValues)

A conditioning case is just a possible combination of values for the parent nodes - a miniature possible world, if you like. The returned distribution must sum to 1, because the entries represent an exhaustive set of cases for the random variable.

Parameters:

parentValues - for the conditioning case. The size of parentValues must equal getParents() and their Random Variables must match.

Returns:

the Probability Distribution for the Random Variable the Conditional Probability Distribution is On.

See Also:

getOn(), getParents()

getSample

java.lang.Object getSample(double probabilityChoice,
                           java.lang.Object... parentValues)

Retrieve a specific example for the Random Variable this conditional distribution is on.

Parameters:

probabilityChoice - a double value, from the range [0.0d, 1.0d), i.e. 0.0d (inclusive) to 1.0d (exclusive).

parentValues - for the conditioning case. The ordering and size of parentValues must equal getParents() and their domains must match.

Returns:

a sample value from the domain of the Random Variable this distribution is on, based on the probability argument passed in.

See Also:

getOn()

getSample

java.lang.Object getSample(double probabilityChoice,
                           AssignmentProposition... parentValues)

Retrieve a specific example for the Random Variable this conditional distribution is on.

Parameters:

probabilityChoice - a double value, from the range [0.0d, 1.0d), i.e. 0.0d (inclusive) to 1.0d (exclusive).

parentValues - for the conditioning case. The size of parentValues must equal getParents() and their Random Variables must match.

Returns:

a sample value from the domain of the Random Variable this distribution is on, based on the probability argument passed in.

See Also:

getOn()