A practice-oriented framework for measuring privacy and utility in data sanitization systems
Bibliographic record
Abstract
Published data is prone to privacy attacks. Sanitization methods aim to prevent these attacks while maintaining usefulness of the data for legitimate users. Quantifying the trade-off between usefulness and privacy of published data has been the subject of much research in recent years. We propose a pragmatic framework for evaluating sanitization systems in real-life and use data mining utility as a universal measure of usefulness and privacy. We propose a definition for data mining utility that can be tuned to capture the needs of data users and the adversaries' intentions in a setting that is specified by a database, a candidate sanitization method, and privacy and utility concerns of data owner. We use this framework to evaluate and compare privacy and utility offered by two well-known sanitization methods, namely k-anonymity and ε-differential privacy, when UCI's "Adult" dataset and the Weka data mining package is used, and utility and privacy measures are defined for users and adversaries. In the case of k-anonymity, we compare our results with the recent work of Brickell and Shmatikov (KDD 2008), and show that using data mining algorithms increases their proposed adversarial gains.
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 | 0.151 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.003 |
| Open science | 0.016 | 0.062 |
| 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".