Revisiting "Privacy Preserving Clustering by Data Transformation"

Abstract

Preserving the privacy of individuals when data are shared for clustering is a complex problem. The challenge is how to protect the underlying data values subjected to clustering without jeopardizing the similarity between objects under analysis. In this short paper, we revisit a family of geometric data transformation methods (GDTMs) that distort numerical attributes by translations, scalings, rotations, or even by the combination of these geometric transformations. Such a method was designed to address privacy-preserving clustering, in scenarios where data owners must not only meet privacy requirements but also guarantee valid clustering results. We offer a detailed, comprehensive and up-to-date picture of methods for privacy-preserving clustering by data transformation.