In k-anonymity, the k is a number that represents the size of a group. If for any individual in the data set, there are at least k-1 individuals who have the same properties, then we have achieved k-anonymity for the data set. Differential privacy (also an industry-standard term) describes a technique for adding mathematical noise to data
(k,ε)-Anonymity: k-Anonymity with ε-Differential Privacy The (k, ϵ)-anonymity approach proposed in this paper first performs k-anonymisation on a subset of the quasi identifiers and then ϵ-differential privacy on the remaining quasi identifiers with different settings for each equivalence class of the k-anonymous dataset. GitHub - llgeek/K-anonymity-and-Differential-Privacy Feb 28, 2019
In the literature, k-anonymization and differential privacy have been viewed as very different privacy guarantees: k-anonymization is syntactic, and differential privacy is algo-rithmic and provides semantic privacy guarantees. Our result is, to our knowledge, the ﬁrst to link k-anonymization with differential privacy.
Mar 17, 2020 Mobile Security and Privacy Over the years, the research community has developed various privacy models, including k-anonymity and differential privacy. In this chapter, we discuss these definitions and implications and the Genetic databases and the future of donor anonymity
Oct 09, 2018
Differential-Private Data Publishing Through Component 1.1 Motivation and Related Work. Data privacy has been an active research topic in statistics, database, and security community for the last three decades [6–9].Many privacy models, such as k-anonymity  and its extensions [11, 12] have been proposed to thwart privacy threats caused by identity and attribute linkages in relational databases.The usual approach is to generalize the records De-identification Tools | NIST