… We have investigated the popularity bias in Graph Collaborative Filtering, and showed it
results from the dimensional collapse of the embedding space. We have demonstrated that the …

… to popularity bias in implicit recommenders to introduce the problem of popularity-opportunity
bias… In this paper, we re-examine popularity bias from the perspective of equal opportunity. …

… , we define what we consider as a popularity bias and how we monitor such a bias throughout
the … This motivated us to complement our measurement of popularity bias with another …

… popularity bias in CF, they often focus on modifying loss functions, which can not fully address
the popularity bias … fundamentally neutralize the popularity bias in the aggregation process …

… bias, the factors impacting popularity bias and its evolution, and the efficacy of methods to
mitigate this bias … This is a common deficiency of collaborative filtering based algorithms and …

… of recommendations: Probabilistic neighborhood selection in collaborative filtering
systems. In Proceedings of the 8th ACM Conference on Recommender systems. 153–160. …

… For this purpose, we select several collaborative filtering algorithms implemented in the
LibRec library [33]: user-based and item-based kNN with cosine similarity [55], BPoissMF [30] (…

… Experiments conducted on four public datasets demonstrate that when equipping two
representative collaborative filtering models with our framework, the popularity bias is effectively …

… of popularity bias [6]: collaborative filtering recommenders typically emphasize popular
items (those with more ratings) over other “long-tail” items [16] that may only be popular among …

… The benefits are twofold: 1) decoupling makes our framework extendable to any collaborative
filtering model that concerns user-item matching; and 2) decoupling enables fast adjusting …