gov.sandia.cognition.learning.algorithm.clustering.divergence

Class GaussianClusterDivergenceFunction

java.lang.Object

gov.sandia.cognition.util.AbstractCloneableSerializable

gov.sandia.cognition.learning.algorithm.clustering.divergence.GaussianClusterDivergenceFunction

All Implemented Interfaces:

ClusterDivergenceFunction<GaussianCluster,Vector>, DivergenceFunction<GaussianCluster,Vector>, CloneableSerializable, java.io.Serializable, java.lang.Cloneable

@CodeReview(reviewer="Kevin R. Dixon",
            date="2008-07-23",
            changesNeeded=false,
            comments="Looks fine.")
public class GaussianClusterDivergenceFunction
extends AbstractCloneableSerializable
implements ClusterDivergenceFunction<GaussianCluster,Vector>

The GaussianClusterDivergenceFunction class implements a divergence function between a Gaussian cluster and a vector, which is calculated by finding the likelihood that the vector was generated from that Gaussian and then returning the negative of the likelihood since it is a divergence measure, not a similarity measure.

Since:

1.0

Author:

Justin Basilico

See Also:

Serialized Form

Field Summary

Fields

Modifier and Type	Field and Description
static GaussianClusterDivergenceFunction	INSTANCE An instance of the class since it has no internal data.

Constructor Summary

Constructors

Constructor and Description
GaussianClusterDivergenceFunction() Creates a new instance of GaussianClusterDivergenceFunction.

Method Summary

Modifier and Type	Method and Description
double	evaluate(GaussianCluster first, Vector second) Evaluates the divergence between the Gaussian cluster and the given vector, which is the negative of likelihood that the cluster was generated.

Methods inherited from class gov.sandia.cognition.util.AbstractCloneableSerializable

clone

Methods inherited from class java.lang.Object

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

Field Detail

INSTANCE

public static final GaussianClusterDivergenceFunction INSTANCE

An instance of the class since it has no internal data.

Constructor Detail

GaussianClusterDivergenceFunction

public GaussianClusterDivergenceFunction()

Creates a new instance of GaussianClusterDivergenceFunction.

Method Detail

evaluate

public double evaluate(GaussianCluster first,
                       Vector second)

Evaluates the divergence between the Gaussian cluster and the given vector, which is the negative of likelihood that the cluster was generated.

Specified by:

evaluate in interface DivergenceFunction<GaussianCluster,Vector>

Parameters:

first - The Gaussian cluster.

second - The vector to calculate the divergence to

Returns:

The divergence between the cluster and vector.