Beyond Centralization: User-Controlled Federated Recommendations in Practice

TL;DR

This paper presents a practical federated recommendation system that enables user control over personalization while preserving privacy, demonstrated through a real-world 53-day user study.

Contribution

It introduces a live federated recommender system allowing user control, with empirical evidence of user preferences and engagement in a real deployment.

Findings

Users prefer personalization when given explicit choice (65.37% CTR).

Participants actively engaged with control mechanisms (average 3.93/5 satisfaction).

Users understood how their interactions influenced recommendations through immediate feedback.

Abstract

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