Utility-preserving anonymization for health data publishing

Lee, Hyukki; Kim, Soo-Hyung; Kim, Jong Wook; Chung, Yon Dohn

doi:10.1186/s12911-017-0499-0

Cited by 61 publications

(52 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…For the multiple numerical sensitive values, the multi sensitive bucketiazation method was proposed based on cluster technology for privacy-preserving data publishing [32]. In addition, several recent approaches [33][34][35] have been proposed to anonymize and publish a dataset while preserving more data utility. However, these methods are vulnerable to composition attack.…”

Section: Related Workmentioning

confidence: 99%

A New Approach to Privacy-Preserving Multiple Independent Data Publishing

et al. 2018

View full text Add to dashboard Cite

Featured Application: The Merging method might apply to publish the datasets sequentially from the different organizations where it will ensure more data utility and privacy. Abstract:We study the problem of privacy preservation in multiple independent data publishing. An attack on personal privacy which uses independent datasets is called a composition attack. For example, a patient might have visited two hospitals for the same disease, and his information is independently anonymized and distributed by the two hospitals. Much of the published work makes use of techniques that reduce data utility as the price of preventing composition attacks on published datasets. In this paper, we propose an innovative approach to protecting published datasets from composition attack. Our cell generalization approach increases both protection of individual privacy from composition attack and data utility. Experimental results show that our approach can preserve more data utility than the existing methods.

show abstract

Section: Related Workmentioning

confidence: 99%

A New Approach to Privacy-Preserving Multiple Independent Data Publishing

et al. 2018

View full text Add to dashboard Cite

show abstract

“…In other words, the adversary examines µ u 's which are defined according to (10) and chooses the only one that belongs to B (n) .…”

Section: Impact Of Dependency On Privacy Usingmentioning

confidence: 99%

Asymptotic Loss in Privacy due to Dependency in Gaussian Traces

Takbiri

Soltani

Goeckel

et al. 2019

2019 IEEE Wireless Communications and Networking Conference (WCNC)

View full text Add to dashboard Cite

The rapid growth of the Internet of Things (IoT) necessitates employing privacy-preserving techniques to protect users' sensitive information. Even when user traces are anonymized, statistical matching can be employed to infer sensitive information. In our previous work, we have established the privacy requirements for the case that the user traces are instantiations of discrete random variables and the adversary knows only the structure of the dependency graph, i.e., whether each pair of users is connected. In this paper, we consider the case where data traces are instantiations of Gaussian random variables and the adversary knows not only the structure of the graph but also the pairwise correlation coefficients. We establish the requirements on anonymization to thwart such statistical matching, which demonstrate the significant degree to which knowledge of the pairwise correlation coefficients further significantly aids the adversary in breaking user anonymity.

show abstract

“…Data sets are anonymized to satisfy certain privacy requirements such as k-anonymity, or l-diversitybefore they are shared with data users. Data anonymization preserves privacy by eliminating the link between people and sensitive information and therefore preventing their identifiability from the dataset (Lee, Kim, Kim, & Chung, 2017) .…”

Section: Data Anonymizationmentioning

confidence: 99%

Feature Based Data Anonymization for High Dimensional Data

Gachanga¹,

Kimwele²,

Nderu³

2019

JIEA

View full text Add to dashboard Cite

Information surges and advances in machine learning tools have enable the collection and storage of large amounts of data. These data are highly dimensional. Individuals are deeply concerned about the consequences of sharing and publishing these data as it may contain their personal information and may compromise their privacy. Anonymization techniques have been used widely to protect sensitive information in published datasets. However, the anonymization of high dimensional data while balancing between privacy and utility is a challenge. In this paper we use feature selection with information gain and ranking to demonstrate that the challenge of high dimensionality in data can be addressed by anonymizing attributes with more irrelevant features. We conduct experiments with real life datasets and build classifiers with the anonymized datasets. Our results show that by combining feature selection with slicing and reducing the amount of data distortion for features with high relevance in a dataset, the utility of anonymized dataset can be enhanced.

show abstract

Utility-preserving anonymization for health data publishing

Cited by 61 publications

References 20 publications

A New Approach to Privacy-Preserving Multiple Independent Data Publishing

A New Approach to Privacy-Preserving Multiple Independent Data Publishing

Asymptotic Loss in Privacy due to Dependency in Gaussian Traces

Feature Based Data Anonymization for High Dimensional Data

Contact Info

Product

Resources

About