From t-Closeness-Like Privacy to Postrandomization via Information Theory

Abstract

Abstract-t-Closeness is a privacy model recently defined for data anonymization. A data set is said to satisfy t-closeness if, for each group of records sharing a combination of key attributes, the distance between the distribution of a confidential attribute in the group and the distribution of the attribute in the entire data set is no more than a threshold t. Here, we define a privacy measure in terms of information theory, similar to t-closeness. Then, we use the tools of that theory to show that our privacy measure can be achieved by the postrandomization method (PRAM) for masking in the discrete case, and by a form of noise addition in the general case.

FSSPCM: Fuzzy Publication of Data for Privacy Preserving

“…To solve this kind of problems, Li et al considered the relationships between global privacy and individual privacy and proposed tcloseness model [7]. In this model, approximation degree of two distributions ha ve been measured by the function of earth mover's distance (EMD), and the differences of distribution between sensitive attributes and the whole published table are required to be no more than t. Reference [8] proposed a complete anonymous algorithm framework based on t-closeness model, which is called SABRE. All of the above methods use generalization technology to meet the privacy requirement of anonymous data.…”

“…Therefore, they proposed K-anonymity method [4][5] to handle the privacy issues during data publication. After that, many scholars have carried out lots of researches based on K-anonymity rules and put forward some improved strategies and anonymous algorithms [6][7][8][9][10][11][12] . The main approach to achieve K-anonymity privacy preserving is generalization, which defines attributes can be associated with others as quasi identifiers.…”

Privacy tradeoffs in predictive analytics

“…To prevent inference of Y from the release, one can apply a distortion f (X) on X; the goal is then to find the minimal distortion so that the mutual information between f (X) and Y is below a threshold. This problem was originally addressed in the asymptotic regime [39,45], while a series of recent works study it in a non-asymptotic setting [7,28,36,38]. Broadly speaking, our work can be cast in this framework by treating a user's ratings as X, her private feature as Y , and employing a correlation structure between them as specified by matrix factorization (namely, (7)).…”

An Information-Theoretic Privacy Criterion for Query Forgery in Information Retrieval

“…However, the problem of k-anonymity, and of enhancements [19][20][21][22] such as l-diversity, is their vulnerability against skewness and similarity attacks [23]. In order to overcome these deficiencies, yet another privacy criterion was considered in [24]: a dataset is said to satisfy t-closeness if for each group of records sharing a combination of key attributes, a certain measure of divergence between the within-group distribution of confidential attributes and the distribution of those attributes for the entire dataset does not exceed a threshold t. An average-case version of the worst-case t-closeness criterion, using the Kullback-Leibler divergence as a measure of discrepancy, turns out to be equivalent to a mutual information, and lend itself to a generalization of Shannon's rate-distortion problem [25,26].…”

“…More recent studies [27,28] rescue the suitable applicability of the concept of entropy as a measure of privacy, by proposing to measure the degree of anonymity observable by an attacker as the entropy of the probability distribution of possible senders of a given message. More recent work has taken initial steps in relating privacy to information-theoretic quantities [3,[24][25][26].…”