Enhancing recommender systems with provider fairness through preference distribution-awareness

Elizabeth Gómez, David Contreras, Ludovico Boratto, Maria Salamó

Abstract: Highlights•Current recommendation models do not support preference distribution-aware provider fairness.•The DAP-fair re-ranking algorithm uses buckets to distribute recommendations.•Recommendations account for both user preferences and provider group representation.•The DAP-fair algorithm was validated on two real-world datasets.•DAP-fair is effective in producing preference distribution-aware provider fairness.