Researchers have developed EdgeRefine, a novel framework designed to enhance the privacy-utility balance in Graph Neural Networks (GNNs). This method addresses the challenge of sensitive link information leakage in graph data by employing adaptive edge refinement and Jaccard similarity for edge probability estimation. EdgeRefine aims to improve accuracy while maintaining strong privacy guarantees, outperforming existing methods by significant margins on benchmark datasets like ACM and Cora. AI
IMPACT Enhances the feasibility of using graph neural networks in privacy-sensitive applications by improving the trade-off between data utility and privacy.
RANK_REASON The cluster contains an academic paper detailing a new method for privacy-preserving graph learning.
- Association for Computing Machinery
- AutoNavi Holdings Limited
- Cora
- EdgeRefine
- graph attention network
- graph convolutional network
- graph neural networks
- Jaccard index
AI-generated summary · Google Gemini · from 2 sources. How we write summaries →
