p20 / Chapter
Federated Learning for Recommender Systems: Advances and Perspectives
Federated Learning Systems: Towards Privacy-Preserving Distributed AI, pp. 87-106 · 2025
In plain language
A review of how federated learning supports personalized recommendations while reducing the need to collect user interactions centrally.
Research context
It offers a starting point for selecting architectures and privacy mechanisms for recommendation services.
Approach
The chapter surveys cryptographic protection, blockchain-based decentralization, and graph-based models for user-item relationships.
Findings
It organizes developments in federated recommendation and discusses remaining limitations and research directions.
Summary based on abstract.