U
- type of the usersI
- type of the itemspublic class AggregateDiversityMetric<U,I> extends EIURD<U,I>
EIURD
multiplied by the cut-off.
S. Vargas. Novelty and diversity evaluation and enhancement in Recommender
Systems. PhD Thesis.
G. Adomavicius and Y. Kwon. Improving aggregate recommendation diversity
using rank-based techniques. TKDE vol. 24 no. 5, 2012.cutoff, freeNorm, itemCount, itemWeight, numUsers
Constructor and Description |
---|
AggregateDiversityMetric(int cutoff,
RelevanceModel<U,I> relModel)
Constructor.
|
Modifier and Type | Method and Description |
---|---|
double |
evaluate()
Evaluates the metric for the recommendations added so far.
|
add, combine, reset
public AggregateDiversityMetric(int cutoff, RelevanceModel<U,I> relModel)
cutoff
- maximum length of the recommendation lists that is evaluatedrelModel
- relevance modelpublic double evaluate()
evaluate
in interface SystemMetric<U,I>
evaluate
in class AbstractSalesDiversityMetric<U,I>
Copyright © 2016. All rights reserved.