Trustworthy Personalized Federated Recommender System with Blockchain-Assisted Decentralized Reward Management

Federated recommender systems (FedRS) enable collaborative model training while preserving user privacy, yet they remain vulnerable to adversarial attacks, unreliable client updates, and misaligned incentives in decentralized environments. Existing approaches struggle to jointly preserve personalization, robustness, and trust when user data are highly non-IID and recommendation quality is governed by ranking-oriented objectives. To address these challenges, we propose a Trustworthy Federated Recommender System (T-FedRS) that extends federated neural collaborative filtering by integrating a ranking-aware reputation mechanism and a lightweight blockchain layer for transparent incentive allocation. Personalization is preserved through locally maintained user embeddings, while item parameters are aggregated using reputation-weighted fusion to suppress malicious updates. Decentralized reputation scores are derived from top-K ranking metrics evaluated on a public probe set and recorded on-chain, enabling verifiable and tamper-resistant reward settlement without exposing raw data or full model updates. Extensive experiments on multiple benchmark datasets demonstrate that T-FedRS consistently outperforms state-of-the-art federated recommender baselines in recommendation accuracy, maintains stable convergence under non-IID data, and remains robust against a high fraction of shilling and model poisoning clients. These findings establish T-FedRS as a practical and effective solution for building trustworthy and personalized FedRS in decentralized settings.