umbc.ebiquity.BayesOWL.coreAlgorithms
Class SDIPFPMarginalOneR

java.lang.Object
  umbc.ebiquity.BayesOWL.coreAlgorithms.SDIPFPMarginalOneR

public class SDIPFPMarginalOneR
extends java.lang.Object

This class implements the 'Simplified D-IPFP' algorithm based on the AISTA-2004 paper, with only one step for one single simple local marginal constraint R(V) provided.

Constructor Summary

SDIPFPMarginalOneR(norsys.netica.Net n, SimpleMarginalConstraint r, HardEvidence[] e)
Constructor.

Method Summary

void	computation() The computation process of one-step simplified de-composed iterative proportional fitting procedure (SDIPFP), for a single simple local marginal constraint.
norsys.netica.Net	getNet() This method returns the revised BBN obtained.

Methods inherited from class java.lang.Object

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

Constructor Detail

SDIPFPMarginalOneR

public SDIPFPMarginalOneR(norsys.netica.Net n,
                          SimpleMarginalConstraint r,
                          HardEvidence[] e)

Constructor.

Parameters:

n: - Bayesian Net

r: - simple marginal constraint

e: - hard evidence

Method Detail

computation

public void computation()

The computation process of one-step simplified de-composed iterative proportional fitting procedure (SDIPFP), for a single simple local marginal constraint.

getNet

public norsys.netica.Net getNet()

This method returns the revised BBN obtained.

Returns:

Bayesian net