es.uam.eps.ir.ranksys.novdiv.distance

Class VectorFeatureItemDistanceModel<I,F>

java.lang.Object

es.uam.eps.ir.ranksys.novdiv.distance.FeatureItemDistanceModel<I,F,Double>

es.uam.eps.ir.ranksys.novdiv.distance.VectorFeatureItemDistanceModel<I,F>

Type Parameters:

I - type of the items

F - type of the features

All Implemented Interfaces:

ItemDistanceModel<I>

Direct Known Subclasses:

CosineFeatureItemDistanceModel, JaccardFeatureItemDistanceModel

public abstract class VectorFeatureItemDistanceModel<I,F>
extends FeatureItemDistanceModel<I,F,Double>

Feature-based item distance model that considers the features of items as vectors.

Author:

Saúl Vargas (saul.vargas@uam.es)

Constructor Summary

Constructors

Constructor and Description
VectorFeatureItemDistanceModel(FeatureData<I,F,Double> featureData) Constructor.

Method Summary

Modifier and Type	Method and Description
protected abstract double	dist(double prod, double norm2A, double norm2B) Distance as a function of the inner product between feature vectors and the square of the norms of these vectors.
ToDoubleFunction<Stream<org.jooq.lambda.tuple.Tuple2<F,Double>>>	dist(Stream<org.jooq.lambda.tuple.Tuple2<F,Double>> features1) Returns a function that returns the feature-based similarity to the features of an item.

Methods inherited from class es.uam.eps.ir.ranksys.novdiv.distance.FeatureItemDistanceModel

dist

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.ranksys.novdiv.distance.ItemDistanceModel

dist

Constructor Detail

VectorFeatureItemDistanceModel

public VectorFeatureItemDistanceModel(FeatureData<I,F,Double> featureData)

Constructor.

Parameters:

featureData - feature data

Method Detail

dist

public ToDoubleFunction<Stream<org.jooq.lambda.tuple.Tuple2<F,Double>>> dist(Stream<org.jooq.lambda.tuple.Tuple2<F,Double>> features1)

Returns a function that returns the feature-based similarity to the features of an item.

Specified by:

dist in class FeatureItemDistanceModel<I,F,Double>

Parameters:

features1 - stream of features of an item

Returns:

function that returns the feature-based similarity to the features of an item

dist

protected abstract double dist(double prod,
                               double norm2A,
                               double norm2B)

Distance as a function of the inner product between feature vectors and the square of the norms of these vectors.

Parameters:

prod - inner product of two vectors

norm2A - square norm of the first vector

norm2B - square norm of the second vector

Returns:

distance value