es.upm.etsisi.cf4j.recommender.knn.userSimilarityMetric

Class CorrelationConstrained

java.lang.Object

es.upm.etsisi.cf4j.recommender.knn.userSimilarityMetric.UserSimilarityMetric

es.upm.etsisi.cf4j.recommender.knn.userSimilarityMetric.CorrelationConstrained

All Implemented Interfaces:

Partible<User>

public class CorrelationConstrained
extends UserSimilarityMetric

Implements traditional Pearson Correlation Constrained as CF similarity metric.

Field Summary

Fields inherited from class es.upm.etsisi.cf4j.recommender.knn.userSimilarityMetric.UserSimilarityMetric

datamodel, similarities

Constructor Summary

Constructors

Constructor and Description
CorrelationConstrained(double median) Constructor of the similarity metric

Method Summary

Modifier and Type	Method and Description
double	similarity(User user, User otherUser) This method must returns the similarity between two users.
String	toString()

Methods inherited from class es.upm.etsisi.cf4j.recommender.knn.userSimilarityMetric.UserSimilarityMetric

afterRun, beforeRun, getSimilarities, run, setDatamodel

Methods inherited from class java.lang.Object

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

Constructor Detail

CorrelationConstrained

public CorrelationConstrained(double median)

Constructor of the similarity metric

Parameters:

median - Median of the ratings of the DataModel instance

Method Detail

similarity

public double similarity(User user,
                         User otherUser)

Description copied from class: UserSimilarityMetric

This method must returns the similarity between two users.

If two users do not have a similarity value, the method must return Double.NEGATIVE_INFINITY.

The value returned by this method should be higher the greater the similarity between users.

Specified by:

similarity in class UserSimilarityMetric

Parameters:

user - A user

otherUser - Other user

Returns:

Similarity between user and otherUser

toString

public String toString()

Overrides:

toString in class UserSimilarityMetric