Flexibly manipulating popularity bias for tackling trade-offs in recommendation

Hiroki Okamura, Keisuke Maeda, Ren Togo, Takahiro Ogawa, Miki Haseyama

Abstract: Highlights•Recommendation models inevitably suffer from trade-offs regarding the popularity of items.•We experimentally reveal a good property of the inner product of recommendation models for long-tailed interaction data.•We can effectively and flexibly manipulate popularity bias simply by manipulating the magnitude of item vectors in our proposed model.