Publications

• Multistakeholder Impacts of Profile Portability in a Recommender Ecosystem

  Anas Buhayh, Elizabeth McKinnie, Clement Canel, Robin Burke

  ACM Conference on User Modeling, Adaptation and Personalization (UMAP), 2026

  Examines how four profile portability policies affect consumers and providers when users can switch between recommendation algorithms. Across three algorithms and three datasets, niche users gained in 35 of 36 scenarios, by up to 192%.

  Paper arXiv Code

• Fairness for Niche Users and Providers: Algorithmic Choice and Profile Portability

  Elizabeth McKinnie, Clement Canel, Anas Buhayh, Robin Burke

  arXiv preprint, 2025

  Simulates algorithmic choice with four profile portability policies on MovieLens data, built on LensKit. Recommendation quality for niche users roughly doubled under every policy.

  arXiv

• Simulating the Algorithm Store: Multistakeholder Impacts of Recommender Choice

  Anas Buhayh, Elizabeth McKinnie, Clement Canel, Robin Burke

  Adjunct Proceedings of ACM UMAP, 2025

  Simulates an algorithm store in which users choose between general and specialized recommenders. Specialized recommenders improved recommendations for niche users by 2.7 to 3.6 times.

  Paper

These are also listed on dblp.