Package es.uam.eps.ir.relison.links.recommendation.metrics.novdiv

Class Unexpectedness<U,​I>

java.lang.Object
es.uam.eps.ir.ranksys.metrics.AbstractSystemMetric<U,​I>
es.uam.eps.ir.relison.links.recommendation.metrics.novdiv.ItemNoveltyMetric<U,​I>
es.uam.eps.ir.relison.links.recommendation.metrics.novdiv.Unexpectedness<U,​I>
Type Parameters:
U - type of the users.
I - type of the items.
All Implemented Interfaces:
es.uam.eps.ir.ranksys.metrics.SystemMetric<U,​I>
public class Unexpectedness<U,​I>
extends ItemNoveltyMetric<U,​I>
Global version of the unexpectedness (expected profile distance).
References:
  1. S. Vargas. Novelty and diversity evaluation and enhancement in Recommender Systems. PhD Thesis (2015)
  2. S. Vargas and P. Castells. Rank and relevance in novelty and diversity for * Recommender Systems. RecSys 2011.
  3. J. Sanz-Cruzado and P. Castells. Beyond Accuracy in Link Prediction. SoMePeaS 2019.

  • Field Summary

    Fields inherited from class es.uam.eps.ir.relison.links.recommendation.metrics.novdiv.ItemNoveltyMetric

    novelty

  • Constructor Summary

    Constructors 
    Constructor Description
    Unexpectedness​(int cutoff, es.uam.eps.ir.ranksys.novelty.unexp.PDItemNovelty<U,​I> distModel)
    Constructor.

  • Method Summary

    Methods inherited from class es.uam.eps.ir.relison.links.recommendation.metrics.novdiv.ItemNoveltyMetric

    add, combine, evaluate, reset

    Methods inherited from class java.lang.Object

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

  • Constructor Details

    • Unexpectedness

      public Unexpectedness​(int cutoff, es.uam.eps.ir.ranksys.novelty.unexp.PDItemNovelty<U,​I> distModel)
      Constructor.
      Parameters:
      cutoff - maximum length of recommendation lists to evaluate
      distModel - item distance model