The publishing of data with privacy guarantees is a task typically performed by a data curator who is expected to provide guarantees for the data he publishes in quantitative fashion, via a privacy criterion (e.g., k-anonymity, l-diversity). The anonymization of data is typically performed off-line. In this paper, we provide algorithmic tools that facilitate the negotiation for the anonymization scheme of a data set in user time. Our method takes as input a set of user constraints for (i) suppression, (ii) generalization and (iii) a privacy criterion (k-anonymity, l-diversity) and returns (a) either an anonymization scheme that fulfils these constraints or, (b) three approximations to the user request based on the idea of keeping the two of the three values of the user input fixed and finding the closest possible approximation for the third parameter. The proposed algorithm involves precomputing suitable histograms for all the different anonymization schemes that a global recoding method can follow. This allows computing exact answers extremely fast (in the order of few milliseconds). © 2012 Springer-Verlag.
CITATION STYLE
Pilalidou, A., & Vassiliadis, P. (2012). Trading privacy for information loss in the blink of an eye. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 7338 LNCS, pp. 572–580). https://doi.org/10.1007/978-3-642-31235-9_38
Mendeley helps you to discover research relevant for your work.