Package es.uam.eps.ir.relison.diffusion.metrics.features.indiv

Class FeatureIndividualKLDivergence<U extends java.io.Serializable,I extends java.io.Serializable,F>

java.lang.Object

es.uam.eps.ir.relison.diffusion.metrics.AbstractIndividualSimulationMetric<U,I,P>

es.uam.eps.ir.relison.diffusion.metrics.features.AbstractFeatureIndividualSimulationMetric<U,I,P>

es.uam.eps.ir.relison.diffusion.metrics.features.indiv.AbstractFeatureKLDivergence<U,I,F>

es.uam.eps.ir.relison.diffusion.metrics.features.indiv.FeatureIndividualKLDivergence<U,I,F>

Type Parameters:

U - type of the users.

I - type of the information pieces.

F - type of the features.

All Implemented Interfaces:

IndividualSimulationMetric<U,I,F>, SimulationMetric<U,I,F>

public class FeatureIndividualKLDivergence<U extends java.io.Serializable,I extends java.io.Serializable,F>
extends AbstractFeatureKLDivergence<U,I,F>

This individual metric computes the number of bytes of information we expect to lose if we approximate the real distribution of features of the users (the total frequency of appearance of the features over the information pieces) with the estimated distribution obtained from simulating. It uses KL Divergence for that. We apply a Laplace smoothing to prevent divisions by zero in both distributions.

Field Summary

Fields

Modifier and Type	Field	Description
private static java.lang.String	ENTROPY	Name fixed value.

Fields inherited from class es.uam.eps.ir.relison.diffusion.metrics.features.indiv.AbstractFeatureKLDivergence

pvalues, qvalues, sumP, sumQ

Fields inherited from class es.uam.eps.ir.relison.diffusion.metrics.AbstractIndividualSimulationMetric

data, initialized

Constructor Summary

Constructors

Constructor	Description
FeatureIndividualKLDivergence(java.lang.String feature, boolean userFeat, boolean unique)	Constructor.

Method Summary

Modifier and Type	Method	Description
double	calculate(U user)	Calculates the metric value for a single user.

Methods inherited from class es.uam.eps.ir.relison.diffusion.metrics.features.indiv.AbstractFeatureKLDivergence

calculate, clear, initialize, updateInfoFeatures, updateUserFeatures

Methods inherited from class es.uam.eps.ir.relison.diffusion.metrics.features.AbstractFeatureIndividualSimulationMetric

getParameter, update, usesUserParam

Methods inherited from class es.uam.eps.ir.relison.diffusion.metrics.AbstractIndividualSimulationMetric

calculateIndividuals, getName, initialize, isInitialized

Methods inherited from class java.lang.Object

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

Methods inherited from interface es.uam.eps.ir.relison.diffusion.metrics.IndividualSimulationMetric

calculate, calculateIndividuals

Methods inherited from interface es.uam.eps.ir.relison.diffusion.metrics.SimulationMetric

calculate, initialize

Field Details

ENTROPY

private static final java.lang.String ENTROPY

Name fixed value.

See Also:

Constant Field Values

Constructor Details

FeatureIndividualKLDivergence

public FeatureIndividualKLDivergence(java.lang.String feature, boolean userFeat, boolean unique)

Constructor.

Parameters:

userFeat - true if we are using a user feature, false if we are using an information piece feature.

feature - the name of the feature.

unique - true if a piece of information is considered once, false if it is considered each time it appears.

Method Details

calculate

public double calculate(U user)

Description copied from interface: IndividualSimulationMetric

Calculates the metric value for a single user.

Parameters:

user - the single user.

Returns:

the value of the metric, NaN if something failed.