Measuring privacy/utility tradeoffs of format-preserving strategies for data release
Bibliographic record
Abstract
In this paper, we introduce a novel approach to evaluate the risk of re-identification of individuals associated with format-preserving data release strategies, focusing on three strategies: data minimization (i.e. through data removal using random sampling and data Shapley values), data anonymization (i.e. through k-anonymity), and data synthesis (i.e. through CTGAN and TVAE generative models). More precisely, our approach consists in simulating a security game in which (1) an attacker performs singling-out attacks as outlined in data protection regulations and (2) an evaluator scores attacks based on the linkability of records and the information gain obtained by the attacker. In addition, we further enhance our approach by simulating attacks as a cooperative game, in which the value of the attackers’ information resources is determined using the Shapley value borrowed from game theory. Re-identification Shapley value is proposed as a method to measure the level of re-identification potential of each feature in a dataset when combined with other features. We demonstrate the effectiveness of our approach using three datasets commonly used in the privacy literature. Overall, our work contributes to a better understanding of the inherent trade-offs that exist between data privacy and data utility in organizations.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.028 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.002 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.004 |
| Open science | 0.038 | 0.039 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; both teacher heads agree on what is shown here.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".