Being accurate is not enough: how accuracy metrics have hurt recommender systems

@article{McNee2006BeingAI,
  title={Being accurate is not enough: how accuracy metrics have hurt recommender systems},
  author={S. McNee and J. Riedl and J. Konstan},
  journal={CHI '06 Extended Abstracts on Human Factors in Computing Systems},
  year={2006}
}
Recommender systems have shown great potential to help users find interesting and relevant items from within a large information space. Most research up to this point has focused on improving the accuracy of recommender systems. We believe that not only has this narrow focus been misguided, but has even been detrimental to the field. The recommendations that are most accurate according to the standard metrics are sometimes not the recommendations that are most useful to users. In this paper, we… Expand

Topics from this paper

A Novel Framework for Improving Recommender Diversity
Beyond rating prediction accuracy: on new perspectives in recommender systems
Improving Aggregate Recommendation Diversity Using Ranking-Based Techniques
Evaluating recommender systems
  • W. Wu, Liang He, J. Yang
  • Computer Science
  • Seventh International Conference on Digital Information Management (ICDIM 2012)
  • 2012
Unifying inconsistent evaluation metrics in recommender systems
Getting recommender systems to think outside the box
...
1
2
3
4
5
...