The classifier judged this item not relevant to AI security, so it is left out of lists, feeds, the API and dataset releases. If that is wrong, report a correction on this page.
InfoResearchPeer-reviewed
NCorr-FP: A Neighbourhood-Based Correlation-Preserving Fingerprinting Scheme for Intellectual Property Protection of Structured Data
- Published
- Record updated
Summary
NCorr-FP is a fingerprinting system for structured tabular data that embeds recipient-specific marks while preserving statistical fidelity. It uses local record similarity and density estimation to guide where fingerprint bits are inserted, then reverses that logic to extract the fingerprint from a possibly modified dataset. Experiments report near-imperceptible fingerprints (minute Hellinger distances and KL divergences), 100% detection confidence under substantial data deletions, and robustness against adaptive and collusion attacks.