What question did this study set out to answer?

To determine whether individuals can be accurately re-identified from incomplete and heavily sampled datasets using a generative statistical framework.

July 23, 2019Nature CommunicationsOpen Access

Estimating the success of re-identifications in incomplete datasets using generative models

View Full Paper

Ask AI

Bookmark

To determine whether individuals can be accurately re-identified from incomplete and heavily sampled datasets using a generative statistical framework.
Developed a generative copula-based model to estimate the likelihood of correctly re-identifying individuals within incomplete datasets.
Evaluated the model's accuracy in predicting individual uniqueness and false-discovery rates across 210 distinct populations.
Achieved AUC scores for predicting individual uniqueness ranging from 0.84 to 0.97 across 210 populations, alongside low false-discovery rates.
Estimated that 99.98% of Americans would be correctly re-identified in any dataset containing 15 demographic attributes.

Discussion

No takes yet. Share an insight, caveat, or question.

Implication

Statistical modeling study reveals that 15 demographic attributes correctly re-identify 99.98% of Americans, indicating current de-identification practices fail modern privacy standards.

Cite This Study

Rocher et al. (2019) studied this question.

synapsesocial.com/papers/6a08ec64ec4e86e9c2e4adef https://doi.org/10.1038/s41467-019-10933-3

View Full Paper

Ask AI

Bookmark

Implication

Statistical modeling study reveals that 15 demographic attributes correctly re-identify 99.98% of Americans, indicating current de-identification practices fail modern privacy standards.

Estimating the success of re-identifications in incomplete datasets using generative models

Key Points

Discussion

Implication

Cite This Study

Estimating the success of re-identifications in incomplete datasets using generative models

Key Points

Discussion

Implication

Cite This Study

Also Consider

Also Consider