Towards Federated Clustering: A Client-wise Private Graph Aggregation Framework

2025-11-14

AI Summary

Federated clustering faces a fundamental trade-off between privacy preservation and clustering performance: transmitting raw embeddings risks exposing sensitive information, while sharing only cluster prototypes degrades accuracy. To address this, we propose SPP-FGC+, a novel framework that leverages client-local structural graphs as privacy-preserving knowledge carriers. It enables global clustering structure modeling without revealing raw data through three core components: private local graph construction, secure aggregation of graph representations, and cross-client graph alignment. Furthermore, we introduce a collaborative feature representation refinement mechanism supporting both one-shot aggregation and iterative optimization. Extensive experiments demonstrate that SPP-FGC+ achieves state-of-the-art performance—improving normalized mutual information (NMI) by up to 10% over existing baselines—while providing provable differential privacy guarantees.

Technology Category

• Machine Learning: Clustering
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization

Application Category

• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• Security and Privacy: Privacy-enhancing technologies
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization

Abstract

Federated clustering addresses the critical challenge of extracting patterns from decentralized, unlabeled data. However, it is hampered by the flaw that current approaches are forced to accept a compromise between performance and privacy: extit{transmitting embedding representations risks sensitive data leakage, while sharing only abstract cluster prototypes leads to diminished model accuracy}. To resolve this dilemma, we propose Structural Privacy-Preserving Federated Graph Clustering (SPP-FGC), a novel algorithm that innovatively leverages local structural graphs as the primary medium for privacy-preserving knowledge sharing, thus moving beyond the limitations of conventional techniques. Our framework operates on a clear client-server logic; on the client-side, each participant constructs a private structural graph that captures intrinsic data relationships, which the server then securely aggregates and aligns to form a comprehensive global graph from which a unified clustering structure is derived. The framework offers two distinct modes to suit different needs. SPP-FGC is designed as an efficient one-shot method that completes its task in a single communication round, ideal for rapid analysis. For more complex, unstructured data like images, SPP-FGC+ employs an iterative process where clients and the server collaboratively refine feature representations to achieve superior downstream performance. Extensive experiments demonstrate that our framework achieves state-of-the-art performance, improving clustering accuracy by up to 10% (NMI) over federated baselines while maintaining provable privacy guarantees.

Problem

• Extracting patterns from decentralized unlabeled data while preserving privacy
• Resolving performance-privacy compromise in federated clustering approaches
• Preventing sensitive data leakage during knowledge sharing in clustering

Innovation

• Uses local structural graphs for privacy-preserving sharing
• Aggregates client graphs into global graph for clustering
• Offers both one-shot and iterative processing modes