Beyond Centralization: User-Controlled Federated Recommendations in Practice
For developers of recommender systems, this work provides the first real-world evidence that user-controlled federated recommendations are feasible and preferred, addressing privacy concerns without sacrificing personalization.
A live federated recommender system was deployed for 53 days with 22 users and 8807 titles, allowing users to control the recommendation objective while keeping data local. Results showed users preferred personalization (65.37% vs 62.07% CTR), rated satisfaction at 3.93/5, and made 248 settings changes, demonstrating that user control, privacy, and effective personalization can be combined.
Recommendation systems typically require centralized user data, limiting user control and raising privacy concerns. Federated learning offers an alternative by keeping data on-device, but its impact on real user behavior remains largely unexplored. We present a live federated recommender system that allows users to control the recommendation objective while keeping their data local. In a 53-day deployment with 22 participants and a catalog of 8807 titles, users interacted with recommendations and switched between personalization and diversity-enhanced ranking. We find that users prefer personalization when given explicit choice (65.37\% vs.\ 62.07\% CTR), actively engage with control mechanisms (3.93/5 satisfaction; 248 settings changes), and develop an understanding of how their interactions affect recommendations through immediate feedback. Our results show that user control, privacy, and effective personalization can be combined in a working system. We demonstrate a practical approach to interactive, privacy-preserving recommendation. Code and demo materials are available at: https://github.com/SlokomManel/federated-recommendations-participants