Exploiting personalized calibration and metrics for fairness recommendation

作者：

Highlights：

• Two new ways to find the personalized trade-off weights.

• Two new metrics to evaluate the calibration in the recommendation list.

• Comparison of three fairness measures in the calibration context.

• Extensive experiments on benchmark composed of six recommender algorithms.

摘要

•Two new ways to find the personalized trade-off weights.•Two new metrics to evaluate the calibration in the recommendation list.•Comparison of three fairness measures in the calibration context.•Extensive experiments on benchmark composed of six recommender algorithms.

论文关键词：Calibration,Collaborative filtering,Fairness,Metrics,Personalization

论文评审过程：Received 6 February 2021, Revised 9 April 2021, Accepted 21 April 2021, Available online 29 April 2021, Version of Record 18 May 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.115112