Street masking: a network-based geographic mask for easily protecting geoprivacy
Bibliographic record
Abstract
BACKGROUND: Geographic masks are techniques used to protect individual privacy in published maps but are highly under-utilized in research. This leads to continual violations of individual privacy, as sensitive health records are put at risk in unmasked maps. New approaches to geographic masking are required that foster accessibility and ease of use, such that they become more widely adopted. This article describes a new geographic masking method, called street masking, that reduces the burden on users of finding supplemental population data by instead automatically retrieving OpenStreetMap data and using the road network as a basis for masking. We compare it to donut geomasking, both with and without population density taken into account, to evaluate its efficacy against geographic masks that require slightly less and slightly more supplemental data. Our analysis is performed on synthetic data in three different Canadian cities. RESULTS: Street masking performs similarly to population-based donut geomasking with regard to privacy protection, achieving comparable k-anonymity values at similar median displacement distances. As expected, distance-based donut geomasking performs worst at privacy protection. Street masking also performs very well regarding information loss, achieving far better cluster preservation and landcover agreement than population-based donut geomasking. Distance-based donut geomasking performs similarly to street masking, though at the cost of reduced privacy protection. CONCLUSION: Street masking competes with, if not out-performs population-based donut geomasking and does so without requiring any supplemental data from users. Moreover, unlike most other geographic masks, it significantly minimizes the risk of false attribution and inherently takes many geographic barriers into account. It is easily accessible for Python users and provides the foundation for interfaces to be built for non-coding users, such that privacy can be better protected in sensitive geospatial research.
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 | 0.001 | 0.004 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
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".