Dale Zimmerman, Department of Statistics, University of Iowa
14 August 2008
The authors did not cite, and therefore appear to be unaware of, an article (of which I am the first author) which overlaps heavily with their article. The article is "Quantifying the effects of mask metadata disclosure and multiple releases on the confidentiality of geographically masked health data," Geographical Analysis, vol. 40, pp. 52-76 (2008). In it we obtain results which are more general than those of the authors, and we obtain these results analytically rather than empirically. In particular we demonstrate what happens to confidentiality (as measured by the area of confidence regions of specified level for individuals' true locations, rather than distance from the average to the true location) as the number of anonymized released versions of the original data set increases. Furthermore, we do this under various scenarios regarding release of metadata describing the anonymization procedures.
A relevant reference
14 August 2008
The authors did not cite, and therefore appear to be unaware of, an article (of which I am the first author) which overlaps heavily with their article. The article is "Quantifying the effects of mask metadata disclosure and multiple releases on the confidentiality of geographically masked health data," Geographical Analysis, vol. 40, pp. 52-76 (2008). In it we obtain results which are more general than those of the authors, and we obtain these results analytically rather than empirically. In particular we demonstrate what happens to confidentiality (as measured by the area of confidence regions of specified level for individuals' true locations, rather than distance from the average to the true location) as the number of anonymized released versions of the original data set increases. Furthermore, we do this under various scenarios regarding release of metadata describing the anonymization procedures.
Competing interests
None declared