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

Local diffusion metrics based on the user and information pieces features.

Class Summary

Class	Description
AbstractExternalFeatureIndividualSimulationMetric<U extends java.io.Serializable,I extends java.io.Serializable,P>	Abstract class representing individual feature-based metrics which do not take into account features that the user already knows (with already knows meaning that the user has the feature, in case of user features, or the user has an information piece containing the feature, in case of information features).
AbstractFeatureKLDivergence<U extends java.io.Serializable,I extends java.io.Serializable,P>	This individual metric computes the KL divergence of the priori distribution of the parameter values over the whole set of information pieces, and the frequency of receival of these parameters for a single user.
ExternalFeatureIndividualGiniComplement<U extends java.io.Serializable,I extends java.io.Serializable,F>	Computes the complement of the Gini coefficient over the distribution of the features that the user does not already know.
ExternalFeatureIndividualRate<U extends java.io.Serializable,I extends java.io.Serializable,F>	Computes the proportion of features that reach a user and are unknown to him/her.
ExternalFeatureRecall<U extends java.io.Serializable,I extends java.io.Serializable,P>	Estimates the fraction of the unknown features of a user have been discovered thanks to the diffusion.
FeatureIndividualEntropy<U extends java.io.Serializable,I extends java.io.Serializable,P>	It computes the entropy of the distribution of times that the different values of a user or information piece feature has reached the different users in the network during a simulation.
FeatureIndividualGini<U extends java.io.Serializable,I extends java.io.Serializable,F>	It computes the Gini complement of the distribution of times that the different values of a user or information piece feature has reached the different users in the network during a simulation.
FeatureIndividualKLDivergence<U extends java.io.Serializable,I extends java.io.Serializable,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.
FeatureKLDivergenceInverse<U extends java.io.Serializable,I extends java.io.Serializable,F>	This individual metric computes the number of bytes of information we expect to lose if we approximate the observed distribution of the parameters received by the user with their prior distribution (i.e.
FeatureRecall<U extends java.io.Serializable,I extends java.io.Serializable,F>	Computes the fraction of all the features that each user has received during the diffusion process.