True spatial k-anonymity: adaptive areal elimination vs. adaptive areal masking

ABSTRACT

Spatial anonymization of address points is critical to fields such as public health. There have been recent concerns about applications of geomasks that did not guarantee the level of k-anonymity theoretically expected. An analysis of the problem and a potential solution were previously proposed: Adaptive Areal Elimination (AAE). The present paper expands on AAE and proposes a modified version, Adaptive Areal Masking (AAM). A benchmark comparison of both methods is conducted, which shows that AAM outperforms AAE in most configurations tested. The discussion attempts to identify the application cases for which AAE might still be preferable and addresses documentation needs with both methods.

KEYWORDS:

• Spatial anonymization
• geomasking
• privacy
• k-anonymity
• spatial aggregation

Disclosure statement

No potential conflict of interest was reported by the authors.

Data Availability

The data and software that support the findings of this study are available at the following permanent links:

• ArcGIS Toolbox: http://doi.org/10.5281/zenodo.3906998

• OGC geopackage of the sample data: http://doi.org/10.5281/zenodo.3906976

Additional information

Funding

The work for this paper was supported by a Pilot Project Grant through the Midwest Center for Occupational Health and Safety (MCOHS) Education and Research Center, University of Minnesota (UMN), Subaward NIOSH T42OH008434.