Research Papers
Robust De-anonymization of Large Sparse Datasets
A study showing how auxiliary information can be used to re-identify records in a sparse dataset.
Why it matters
Anonymisation depends on the surrounding information environment. A dataset that looks anonymous on its own can become identifying when joined with another dataset.
Claim labels
FACT
The paper was presented at the 2008 IEEE Symposium on Security and Privacy and is identified by DOI 10.1109/SP.2008.33.
RESEARCHER ANALYSIS
The paper studies a particular class of datasets and attack assumptions; it is not evidence that every anonymisation method fails in the same way.
Sources
Last link check: 2026-09-27. A link check confirms reachability, not that every claim has been independently reviewed.