MaskMyPy: python tools for performing and analyzing geographic masks
Bibliographic record
Abstract
BACKGROUND: Geographic masking is an important but under-utilized technique for protecting and disseminating sensitive geospatial health data. Geographic masks work by displacing static point locations such that the people those locations describe cannot be identified, while at the same time preserving important spatial patterns for analysis. Unfortunately, there is a lack of available tooling surrounding geographic masks which we believe creates an unnecessary barrier towards the adoption of these techniques. As such, this article presents a set of tools for performing, evaluating, and developing geographic masks, called MaskMyPy. RESULTS: MaskMyPy is an open-source Python package that includes functions for performing geographic masks, including donut, street, location swapping, and Voronoi masks. It also includes a range of tools for evaluating the results of these masks, both with regard to privacy and information loss. Finally, it includes a special class called the 'Atlas' that aims to dramatically streamline mask execution and evaluation. We conducted a short case study to illustrate the power of MaskMyPy in geographic masking research, and in doing so showed that mask performance can range widely due solely to randomization. As such, we recommend that masking researchers test their masks repeatedly across a variety of test datasets. CONCLUSION: MaskMyPy makes it easy to apply a variety of geographic masks to a set of sensitive points and then measure which mask provided the most privacy while suffering the least information loss. We believe this style of tooling is important to not only make geographic masks accessible to non-experts, but to enable expert users to better interrogate the masks they develop, and in doing so drive the geographic masking discipline forward.
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 machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.003 | 0.016 |
| Meta-epidemiology (narrow) | 0.003 | 0.002 |
| Meta-epidemiology (broad) | 0.001 | 0.003 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.001 | 0.002 |
| Scholarly communication | 0.003 | 0.005 |
| Open science | 0.004 | 0.006 |
| Research integrity | 0.001 | 0.004 |
| Insufficient payload (model declined to judge) | 0.036 | 0.024 |
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; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
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".