edu.ksu.cis.bnj.bbn.inference.ls
Class LS
java.lang.Object
edu.ksu.cis.bnj.bbn.inference.Inference
edu.ksu.cis.bnj.bbn.inference.ExactInference
edu.ksu.cis.bnj.bbn.inference.ls.LS
- All Implemented Interfaces:
- Optionable
- public class LS
- extends ExactInference
Lauritzen-Spiegelhalter algorithm.
This algorithm is divided into two main phase: Constructing the clique tree
and propagating the belief.
For clique tree construction, please see CliqueTree.java at
edu.ksu.cis.kdd.util.graph package (not at this package)
For propagation, see Clique.java at this package.
I decided to split these implementation because Sparse Candidate also need
clique tree but of different representation. This package's CliqueTree simply
"repackage" the result of clique tree algorithm.
- Author:
- Roby Joehanes
Methods inherited from class edu.ksu.cis.bnj.bbn.inference.Inference |
execute, getCurrentOptions, getDefaultOptions, getGraph, getMAP, getMarginalsResult, getMPE, getOptionsDialog, getOutputFile, getRunType, load, setGraph, setOption, setOptions, setOutputFile, setRunType |
Methods inherited from class java.lang.Object |
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait |
tree
protected CliqueTree tree
LS
public LS()
LS
public LS(BBNGraph g)
- Constructor for LS.
- Parameters:
g
-
getName
public java.lang.String getName()
- Specified by:
getName
in class Inference
getMarginals
public InferenceResult getMarginals()
- Description copied from class:
Inference
- Getting the result of the marginals (i.e. the belief revision)
- Specified by:
getMarginals
in class Inference
- Returns:
- InferenceResult The marginals
- See Also:
Inference.getMarginals()
main
public static void main(java.lang.String[] args)