Measuring and Mitigating Item Under-Recommendation Bias in Personalized Ranking Systems
Measuring and Mitigating Item Under-Recommendation Bias in Personalized Ranking Systems
复制标题
DOI：
10.1145/3397271.3401177
复制
发表时间：
2020-07
期刊：
影响因子：
--
通讯作者：
Ziwei Zhu;Jianling Wang;James Caverlee
中科院分区：
文献类型：
--
作者：
Ziwei Zhu;Jianling Wang;James Caverlee
推荐算法通常基于用户-项目交互(例如，点击、喜欢或评级)来构建模型，以提供个性化的项目排名列表。由于不同的用户偏好，这些交互通常不均匀地分布在不同的项目组上。然而，我们发现，推荐算法可以继承甚至放大这种不平衡的分布，从而导致项目推荐不足的偏差。具体地说，我们将基于排名的统计平等性和机会均等的概念形式化，作为衡量项目推荐不足偏差的两个度量。然后，我们的经验表明，最广泛采用的算法之一--贝叶斯个性化排名--会产生有偏见的推荐，这促使我们努力提出新的无偏见个性化排名模型。该去偏模型能够在保持推荐性能的同时改善两个提出的偏向度量。在三个公开数据集上的实验表明，与最先进的替代方案相比，所提出的模型具有很强的偏倚减少。
Recommendation algorithms typically build models based on user-item interactions (e.g., clicks, likes, or ratings) to provide a personalized ranked list of items. These interactions are often distributed unevenly over different groups of items due to varying user preferences. However, we show that recommendation algorithms can inherit or even amplify this imbalanced distribution, leading to item under-recommendation bias. Concretely, we formalize the concepts of ranking-based statistical parity and equal opportunity as two measures of item under-recommendation bias. Then, we empirically show that one of the most widely adopted algorithms -- Bayesian Personalized Ranking -- produces biased recommendations, which motivates our effort to propose the novel debiased personalized ranking model. The debiased model is able to improve the two proposed bias metrics while preserving recommendation performance. Experiments on three public datasets show strong bias reduction of the proposed model versus state-of-the-art alternatives.