Package com.bayesserver.analysis

Class DSeparation

java.lang.Object

com.bayesserver.analysis.DSeparation

public final class DSeparation
extends Object

Contains methods to calculate D-Separation.

Method Summary

Modifier and Type	Method	Description
static DSeparationOutput	calculate(Network network, List<Node> sourceNodes, List<Node> testNodes, Evidence evidence, DSeparationOptions options)	Calculates whether sets of nodes are D-Separated, given any evidence.
static DSeparationOutput	calculate(Network network, List<Node> sourceNodes, List<Integer> sourceNodeTimes, List<Node> testNodes, List<Integer> testTimes, Evidence evidence, DSeparationOptions options)	Calculates whether sets of nodes are D-Separated, given any evidence, and associated times for any temporal nodes.

Methods inherited from class java.lang.Object

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

Method Detail

calculate

public static DSeparationOutput calculate(Network network,
                                          List<Node> sourceNodes,
                                          List<Node> testNodes,
                                          Evidence evidence,
                                          DSeparationOptions options)

Calculates whether sets of nodes are D-Separated, given any evidence.

Parameters:

network - The Bayesian network.

sourceNodes - A set of nodes, which if changed (new evidence set), may or may not affect the test nodes.

testNodes - The nodes we are testing to see if they would be effected by new evidence on any of the source nodes.

evidence - Any existing evidence.

options - Options governing the calculations.

Returns:

An instance which contains the d-separation results for each test node.

calculate

public static DSeparationOutput calculate(Network network,
                                          List<Node> sourceNodes,
                                          List<Integer> sourceNodeTimes,
                                          List<Node> testNodes,
                                          List<Integer> testTimes,
                                          Evidence evidence,
                                          DSeparationOptions options)

Calculates whether sets of nodes are D-Separated, given any evidence, and associated times for any temporal nodes.

Parameters:

network - The Bayesian network.

sourceNodes - A set of nodes, which if changed (new evidence set), may or may not affect the test nodes.

sourceNodeTimes - Any times associated with the test nodes.

testNodes - The nodes we are testing to see if they would be effected by new evidence on any of the source nodes.

testTimes - Any times associated with the test nodes.

evidence - Any existing evidence.

options - Options governing the calculations.

Returns:

An instance which contains the d-separation results for each test node.