unbbayes.prs.hybridbn
Class GaussianMixture

java.lang.Object
  unbbayes.prs.hybridbn.GaussianMixture

All Implemented Interfaces:

IInferenceAlgorithm

public class GaussianMixture
extends Object
implements IInferenceAlgorithm

Gaussian Mixture inference algorithm. It calculates the prior (compilation) for the discrete nodes and the mean and variance for the continuous nodes. The propagation is not yet implemented.

Author:

Rommel Carvalho

Field Summary

protected ProbabilisticNetwork	clonedPN
protected List<Node>	nodeOrderQueue
protected ProbabilisticNetwork	pn

Constructor Summary

GaussianMixture()
Default constructor - created for plugin support

GaussianMixture(ProbabilisticNetwork pn)

Method Summary

protected void	addAdjacentNodes(Node node, List<Node> nodeInNetwork)
protected void	addToOrderQueue(ArrayList<Node> children, boolean[] nodeAddedList) Take the children of a node that have already been added to the queue.
protected ProbabilisticNetwork	clonePN(ProbabilisticNetwork network) As I am not sure if the clone methods are corrected.
protected void	createOrderQueue() Creates the queue of the nodes that are going to be analyzed.
String	getDescription() Obtains the description of this algorithm.
protected Integer	getIndexInQueue(Node node) Retrieves the node's index in the queue.
String	getName() Obtains the name of this algorithm
Graph	getNetwork()
List<Node>	getNodeOrderQueue() Return the order the nodes are in the sampled matrix.
protected List<Integer>	getParentsIndexesInQueue(ProbabilisticNode node) Return the indexes (sampling order) in the queue for the parents of a given node.
protected void	init()
protected void	initOrderQueue(boolean[] nodeAddedList) Initializes the queue with the nodes that are root.
void	propagate() Propagate evidences.
void	reset() Resets the algorithm and optionally the network state.
void	run() Runs the algorithm given the current state of the attributes.
void	setNetwork(Graph g) Sets the network (graph) treated by this algorithm.

Methods inherited from class java.lang.Object

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

Field Detail

pn

protected ProbabilisticNetwork pn

clonedPN

protected ProbabilisticNetwork clonedPN

nodeOrderQueue

protected List<Node> nodeOrderQueue

Constructor Detail

GaussianMixture

public GaussianMixture()

Default constructor - created for plugin support

GaussianMixture

public GaussianMixture(ProbabilisticNetwork pn)

Method Detail

getNodeOrderQueue

public List<Node> getNodeOrderQueue()

Return the order the nodes are in the sampled matrix.

Returns:

The order the nodes are in the sampled matrix.

init

protected void init()

createOrderQueue

protected void createOrderQueue()

Creates the queue of the nodes that are going to be analyzed.

initOrderQueue

protected void initOrderQueue(boolean[] nodeAddedList)

Initializes the queue with the nodes that are root. In other words. It will put in the queue the nodes that do not have parents.

Parameters:

nodeAddedList - Keeps track of the nodes that have already been added to the queue (nodeAddedList[nodeIndex]=true).

addToOrderQueue

protected void addToOrderQueue(ArrayList<Node> children,
                               boolean[] nodeAddedList)

Take the children of a node that have already been added to the queue. Analyze them one by one and add the child that is not in the queue yet.

Parameters:

children - Children of a node that is already in the queue.

nodeAddedList - Nodes that have already been added to the queue.

run

public void run()
         throws IllegalStateException

Description copied from interface: IInferenceAlgorithm

Runs the algorithm given the current state of the attributes.

Specified by:

run in interface IInferenceAlgorithm

Throws:

IllegalStateException - : this exception may also be used for general purpose.

addAdjacentNodes

protected void addAdjacentNodes(Node node,
                                List<Node> nodeInNetwork)

getParentsIndexesInQueue

protected List<Integer> getParentsIndexesInQueue(ProbabilisticNode node)

Return the indexes (sampling order) in the queue for the parents of a given node.

Parameters:

node - The node to retrieve the parents for finding the indexes.

Returns:

List of indexes (sampling order) of a node's parents in the queue.

getIndexInQueue

protected Integer getIndexInQueue(Node node)

Retrieves the node's index in the queue.

Parameters:

node -

Returns:

clonePN

protected ProbabilisticNetwork clonePN(ProbabilisticNetwork network)

As I am not sure if the clone methods are corrected. I decided to clone the network by saving it in a file and loading again as another network.

Parameters:

network -

Returns:

setNetwork

public void setNetwork(Graph g)
                throws IllegalArgumentException

Description copied from interface: IInferenceAlgorithm

Sets the network (graph) treated by this algorithm.

Specified by:

setNetwork in interface IInferenceAlgorithm

Parameters:

g - : the network (graph) to set.

Throws:

IllegalArgumentException - : if the given graph is not appropriate for this algorithm.

getNetwork

public Graph getNetwork()

Specified by:

getNetwork in interface IInferenceAlgorithm

Returns:

Gets the network (graph) treated by this algorithm.

getDescription

public String getDescription()

Description copied from interface: IInferenceAlgorithm

Obtains the description of this algorithm. This information is generally used as a tool tip text.

Specified by:

getDescription in interface IInferenceAlgorithm

Returns:

: name of this algorithm.

getName

public String getName()

Description copied from interface: IInferenceAlgorithm

Obtains the name of this algorithm

Specified by:

getName in interface IInferenceAlgorithm

Returns:

: name of this algorithm.

reset

public void reset()

Description copied from interface: IInferenceAlgorithm

Resets the algorithm and optionally the network state.

Specified by:

reset in interface IInferenceAlgorithm

propagate

public void propagate()

Description copied from interface: IInferenceAlgorithm

Propagate evidences.

Specified by:

propagate in interface IInferenceAlgorithm