edu.ucla.sspace.similarity

Class KLDivergence

java.lang.Object

edu.ucla.sspace.similarity.KLDivergence

All Implemented Interfaces:

SimilarityFunction

public class KLDivergence
extends Object
implements SimilarityFunction

Returns the KL Divergence between any two probability distributions represented as Vectors.

This metric is not symmetric. Use the Jensen-Shannon Divergence for a symmetric similarity measure between two probability distributions

Author:

Keith Stevens

Constructor Summary

Constructors

Constructor and Description
KLDivergence()

Method Summary

Methods

Modifier and Type	Method and Description
boolean	isSymmetric() Returns false (is asymmetric).
void	setParams(double... arguments) Does nothing
double	sim(DoubleVector v1, DoubleVector v2) Returns the similarity between v1 and v2.
double	sim(IntegerVector v1, IntegerVector v2) Returns the similarity between v1 and v2.
double	sim(Vector v1, Vector v2) Returns the similarity between v1 and v2.

Methods inherited from class java.lang.Object

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

Constructor Detail

KLDivergence

public KLDivergence()

Method Detail

setParams

public void setParams(double... arguments)

Does nothing

Specified by:

setParams in interface SimilarityFunction

isSymmetric

public boolean isSymmetric()

Returns false (is asymmetric).

Specified by:

isSymmetric in interface SimilarityFunction

sim

public double sim(DoubleVector v1,
         DoubleVector v2)

Returns the similarity between v1 and v2. If SimilarityFunction.isSymmetric() is false, the ordering does matter.

Specified by:

sim in interface SimilarityFunction

sim

public double sim(IntegerVector v1,
         IntegerVector v2)

Returns the similarity between v1 and v2. If SimilarityFunction.isSymmetric() is false, the ordering does matter.

Specified by:

sim in interface SimilarityFunction

sim

public double sim(Vector v1,
         Vector v2)

Returns the similarity between v1 and v2. If SimilarityFunction.isSymmetric() is false, the ordering does matter.

Specified by:

sim in interface SimilarityFunction