An entropy based method for measuring anonymity

Bezzi, Michele

doi:10.1109/seccom.2007.4550303

Cited by 26 publications

(32 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…statistical disclosure control protection methods [49]) is applied only to the non-confidential attributes leaving the confidential attributes. Attributes based disclosure risk estimations attempt [34,35], Kernel distance, Choquet integral based distance [36]), probabilistic record linkage metrics [37] To identify the link between data Similarity based membership k-anonymity [38], l-diversity [39] and t-closeness [40] To reveal the existence of sensitive attributes Entropy Entropy based metrics [41][42][43][44][45] To quantify the extent of information loss Statistical disclosure control aggregation [46], rounding [47], swapping [48], adding random noise to data…”

Section: Disclosure Risk Measuresmentioning

confidence: 99%

Location-dependent disclosure risk based decision support framework for persistent authentication in pervasive computing applications

2015

View full text Add to dashboard Cite

Section: Disclosure Risk Measuresmentioning

confidence: 99%

Location-dependent disclosure risk based decision support framework for persistent authentication in pervasive computing applications

2015

View full text Add to dashboard Cite

“…Mutual information has been shown as an effective screen-space metric for parallel coordinates [9]. In the privacy context, we use lack of mutual information as a measure of high uncertainty [4]. If there is high mutual information between two axes like creditamount and credithistory (Figure 11), there might be a skewed distribution which makes it easier for an intruder to breach the privacy if he has some background knowledge about the person.…”

Section: Different Reordering Configurationsmentioning

confidence: 99%

Adaptive Privacy-Preserving Visualization Using Parallel Coordinates

Dasgupta

Kosara

2011

IEEE Trans. Visual. Comput. Graphics

View full text Add to dashboard Cite

Abstract-Current information visualization techniques assume unrestricted access to data. However, privacy protection is a key issue for a lot of real-world data analyses. Corporate data, medical records, etc. are rich in analytical value but cannot be shared without first going through a transformation step where explicit identifiers are removed and the data is sanitized. Researchers in the field of data mining have proposed different techniques over the years for privacy-preserving data publishing and subsequent mining techniques on such sanitized data. A well-known drawback in these methods is that for even a small guarantee of privacy, the utility of the datasets is greatly reduced. In this paper, we propose an adaptive technique for privacy preservation in parallel coordinates. Based on knowledge about the sensitivity of the data, we compute a clustered representation on the fly, which allows the user to explore the data without breaching privacy. Through the use of screen-space privacy metrics, the technique adapts to the user's screen parameters and interaction. We demonstrate our method in a case study and discuss potential attack scenarios.

show abstract

“…Claude Shannons seminal paper: "A Mathematical Theory of Communication" was the first paper proving that the level of ambiguity or equivocation in the transferred information can be measured using an entropy-based metric [18]. Shannon entropy is also proposed as a metric of anonymity by [1,16]. Rényi entropy is proposed as a more general metric of anonymity that can express several degrees of entropies from Max-entropy, Shannon entropy and to Min-entropy as a limiting case when the degree α → ∞ [3].…”

Section: Related Workmentioning

confidence: 99%

A Composite Privacy Leakage Indicator

Ulltveit-Moe

Oleshchuk

2011

Wireless Pers Commun

View full text Add to dashboard Cite

This paper proposes a Subjective Logic based composite privacy leakage metric that both takes into account the amount of information leakage and also that information with high entropy in some cases may be considered encrypted. It is furthermore shown both analytically and experimentally that Min-entropy is considered better than Shannon, Rényi or Max entropy for identifying encrypted content for the composite metric. This is in particular useful for implementing privacy-enhanced Intrusion Detection Systems (IDS), where sampled encrypted traffic can be considered to have low risk of revealing sensitive information. The combined metric can be used in a Policy Enforcement Point that acts as a proxy/anonymiser in order to to reduce the leakage of private or sensitive information from the IDS sensors to an outsourced Managed Security Service provider. Although the composite privacy indicator is IDS specific, the authorisation architecture is general, and may also be useful for anonymising or pseusonymising sensitive information from or to other types of sensors that need to be exposed to the Internet. The solution is based on the eXtensible Access Control Markup Language policy language extended with support for Subjective Logic, in order to provide a method for expressing fine-grained access control policies that are based on uncertain evidences.

show abstract

An entropy based method for measuring anonymity

Cited by 26 publications

References 12 publications

Location-dependent disclosure risk based decision support framework for persistent authentication in pervasive computing applications

Location-dependent disclosure risk based decision support framework for persistent authentication in pervasive computing applications

Adaptive Privacy-Preserving Visualization Using Parallel Coordinates

A Composite Privacy Leakage Indicator

Contact Info

Product

Resources

About