Corpus ID: 42914272

The Demographics of Cool: Popularity and Recommender Performance for Different Groups of Users

@inproceedings{Ekstrand2017TheDO,
  title={The Demographics of Cool: Popularity and Recommender Performance for Different Groups of Users},
  author={Michael D. Ekstrand and M. S. Pera},
  booktitle={RecSys Posters},
  year={2017}
}
ABSTRACT Typical recommender evaluations treat users as an homogeneous unit. However, user subgroups often differ in their tastes, which can result more broadly in diverse recommender needs. Thus, these groups may have different degrees of satisfaction with the provided recommendations. We explore the offline top-N performance of collaborative filtering algorithms across two domains. We find that several strategies achieve higher accuracy for dominant demographic groups, thus increasing the… Expand
Per-Instance Algorithm Selection for Recommender Systems via Instance Clustering
A Novel Approach to Recommendation Algorithm Selection using Meta-Learning
Contextual Meta-Bandit for Recommender Systems Selection

References

SHOWING 1-8 OF 8 REFERENCES
Precision-oriented evaluation of recommender systems: an algorithmic comparison
Performance of recommender algorithms on top-n recommendation tasks
Evaluating Recommendation Systems
The MovieLens Datasets: History and Context