Name
Abstract | ||
|---|---|---|
Interval inference is a specific kind of statisticaldisclosure where a snooper collects and analyzes publiclyavailable data to determine tight bounds on confidentialnumerical data. Institutions that disseminate public datainclude Census Bureaus and other independentorganizations such as regional healthcare initiatives thatprovide chronic disease data that is collected fromphysicians, pharmacies and health maintenanceorganizations (HMOs). Such initiatives must ensure thatthe confidential values of the data providers are protectedagainst interval inference while making sure that thereleased information is still useful for the prospectivedata users (such as medical researchers). In this paper,we consider the important case of 2-dimensional tableswhere the rows correspond to the data providers and thecolumns to confidential data categories. Although theinner cells of this table are confidential and should underno circumstances be published, marginal informationabout central tendency and dispersion can still be usefuland worth publishing. It is the task of the data-disseminatinginstitution to elicit these specific marginaldata elements for publication such that no tight bounds onany inner table cell can be inferred. We present a newmethod that maximizes the usefulness of the disseminatedinformation to the prospective data users while ensuringthe confidentiality of the inner table cell values. We give acomputational analysis and compare our methods toexisting statistical disclosure methods. |
Year | DOI | Venue |
|---|---|---|
2004 | 10.1109/SSDBM.2004.29 | SSDBM |
Keywords | Field | DocType |
confidentialnumerical data,tight bound,data provider,inner table cell,interval inference,prospective data user,chronic disease data,statistical databases,confidential data category,thatthe confidential value,inner table cell value,analyzes publiclyavailable data,data privacy,census,databases,data analysis,data confidentiality,data dissemination,2 dimensional,public health | Row,Health care,Data science,Data mining,Confidentiality,Computer science,Inference,Dissemination,Information Dissemination,Publishing,Information privacy,Database | Conference |
ISBN | Citations | PageRank |
0-7695-2146-0 | 0 | 0.34 |
References | Authors | |
8 | 2 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Claus Boyens | 1 | 37 | 4.77 |
| Oliver Günther | 2 | 85 | 29.91 |