aima.core.probability.bayes

Interface ConditionalProbabilityTable

All Superinterfaces:

ConditionalProbabilityDistribution

All Known Implementing Classes:

CPT

public interface ConditionalProbabilityTable
extends ConditionalProbabilityDistribution

Artificial Intelligence A Modern Approach (3rd Edition): page 512.

A Conditional Probability Table, or CPT, can be used for representing conditional probabilities for discrete (finite) random variables. Each row in a CPT contains the conditional probability of each node value for a conditioning case.

Author:

Ciaran O'Reilly

Method Summary

Modifier and Type	Method and Description
CategoricalDistribution	getConditioningCase(AssignmentProposition... parentValues) A conditioning case is just a possible combination of values for the parent nodes - a miniature possible world, if you like.
CategoricalDistribution	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.
Factor	getFactorFor(AssignmentProposition... evidence) Construct a Factor consisting of the Random Variables from the Conditional Probability Table that are not part of the evidence (see AIMA3e pg.

Methods inherited from interface aima.core.probability.bayes.ConditionalProbabilityDistribution

contains, getFor, getOn, getParents, getSample, getSample, getValue, getValue

Method Detail

getConditioningCase

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

Description copied from interface: ConditionalProbabilityDistribution

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.

Specified by:

getConditioningCase in interface ConditionalProbabilityDistribution

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:

ConditionalProbabilityDistribution.getOn(), ConditionalProbabilityDistribution.getParents()

getConditioningCase

CategoricalDistribution getConditioningCase(AssignmentProposition... parentValues)

Description copied from interface: ConditionalProbabilityDistribution

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.

Specified by:

getConditioningCase in interface ConditionalProbabilityDistribution

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:

ConditionalProbabilityDistribution.getOn(), ConditionalProbabilityDistribution.getParents()

getFactorFor

Factor getFactorFor(AssignmentProposition... evidence)

Construct a Factor consisting of the Random Variables from the Conditional Probability Table that are not part of the evidence (see AIMA3e pg. 524).

Parameters:

evidence -

Returns:

a Factor for the Random Variables from the Conditional Probability Table that are not part of the evidence.