Achieving anonymity via clustering

By Gagan Aggarwal, Tomás Feder, Krishnaram Kenthapadi, Samir Khuller, Rina Panigrahy, Dilys Thomas, An Zhu. ACM SIGACT-SIGMOD-SIGART Symposium on Principles of Database Systems, 2006

405 citations recorded by the source corpus.

Subfields: Artificial Intelligence

View Paper: https://www.semanticscholar.org/paper/5353f30d4c015a0e9e8b57718cd6525612cb5a66

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

Citations in the corpus

On the cost of essentially fair clusterings (2018, background): “The wide applicability of k -center has given rise to numerous variants that strive to incorporate useful con-19 straints into the solution, such as capacity constraints [10, 15, 24], lower bounds on the size of each cluster [3, 6] or allowing for outliers [12, 16].”
Privacy preserving clustering with constraints (2018, background): “In private clustering [2], we demand a lower bound on the number of points assigned to a center to ensure a certain anonymity.”
Privacy-preserving anonymization of set-valued data (2008, methodology): “[1] model the problem as clustering and propose a constant factor approximation of the optimal solution, but the bound only holds for the Euclidean distance metric.”
Never Walk Alone: Uncertainty for Anonymity in Moving Objects Databases (2008, background): “It is worth noting that publishing one unique representative per cluster, for a number of times equal to the population of the cluster, can be seen as a special instance (δ = 0, i.e., no uncertainty) of our framework.”
Fair Algorithms for Clustering (2019, background): “The problem arises when, for example, one wants to ensure anonymity [4] and is called “pri-vate clustering” in [37].”
Utility-Privacy Tradeoffs in Databases: An Information-Theoretic Approach (2011, background): “s shown that approximation algorithms that can run on original histograms can also run on the sanitized histograms with a degradation of performance. Clustering, a common sanitization technique [17], [40], [41], is claimed to maintain utility as a result of the following property: all points in a cl”
Towards trajectory anonymization: a generalization-based approach (2008, background): “Most clustering algorithms can easily be modified for k-anonymity by enforcing that the size of the clusters should be more than k [2, 38, 14, 1].”
Privacy Preservation in the Publication of Trajectories (2008, methodology): “In [1], k-anonymity is treated as a special clustering problem, and a constant factor approximation of the optimal solution is proposed.”
Application of Local Differential Privacy to Collection of Indoor Positioning Data (2018, background): “For example, [1], [2] group k similar records into a cluster and generalize each cluster to achieve k -anonymity.”
A General Survey of Privacy-Preserving Data Mining Models and Algorithms (2008, methodology): “The work in [15] also provides constant factor approximation algorithms to design the clustering.”
Privacy-preserving data publishing: A survey of recent developments (2010, background): “…Top-Down Greedy [Xu et al. 2006] CG DM minimal TDS2P [Wang et al. 2005; Mohammed et al. 2009] SG IGPL minimal Condensation [Aggarwal and Yu 2008a, 2008b] CD heuristics minimal r-Gather Clustering [Aggarwal et al. 2006] CL heuristics minimal Attribute Linkage Top-Down Disclosure [Wang et al.”
Fast Data Anonymization with Low Information Loss (2007, background): “[1] k-anonymity is treated as a special clustering problem, called r-cellular clustering.”