Linkage Attack

Linkage attacks exploit the relationships between data points across different datasets to infer sensitive information, posing a significant threat to privacy in various applications using machine learning models. Current research focuses on mitigating these attacks through techniques like data fragmentation, differential privacy, and the development of novel model architectures such as privacy-preserving graph neural networks (GNNs) that limit the inference of sensitive links within the data. This research is crucial for ensuring the responsible use of data in sensitive domains like healthcare and finance, where the risk of re-identification and privacy breaches is high.