Disclosure risk assessment via record linkage by a maximum-knowledge attacker

Domingo-Ferrer, Josep; Ricci, Sara

doi:10.1109/pst.2015.7232951

Cited by 27 publications

(19 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This masked version is then evaluated in terms of disclosure risk and information loss according to currently existing metrics, which are generally specific to the environment. In the permutation paradigm, on the right of Figure 1 4 , permutation matrices are identified from the anonymized data set, from which can be retrieved, as will be developed below, vectors of rank displacements and relative rank displacements, which can in turn be used to compute universal measures of disclosure risk and information loss. We qualify these measures as universal for two reasons.…”

Section: A New Roadmap For Data Anonymizationmentioning

confidence: 99%

“…They are not specific to the environment. The second is that they can 4 Note that for the same reasons outlined above, permutation matrices can potentially be considered as a direct input for anonymization, replacing the choice of method. incorporate different judgments on disclosure risk and information loss.…”

Section: A New Roadmap For Data Anonymizationmentioning

confidence: 99%

“…It allows nonetheless in a simple fashion to do a first screening of protection against disclosure risk. In the example of Table 2, ) =0, 1 =0.6 and 2 =0.8; of the three attributes, the first has no protection against disclosure risk while the third has the highest protection, according to 4 .…”

Section: A Class Of Universal Measures Of Disclosure Risk Based On Pementioning

confidence: 99%

“…where r j(i) denotes the elements of r j and α the parameter of aversion to disclosure risk. 4 (α) makes use of a power mean 5 (see [7] for a discussion of its various properties) for the aggregation of the components of r j , with the parameter substantiating the notion of aversion to disclosure risk. The arithmetic mean becomes a special case (α = 1) of 4 (α), which forms a natural starting point by computing the average level of permutation distances.…”

Section: A Class Of Universal Measures Of Disclosure Risk Based On Pementioning

confidence: 99%

“…From these proposals more measures can be elaborated. For example, one could think in combining 4 and 4 (1) to get: 4 = 4 * 4 (1) 4 measures the average permutation distances for attribute j, but discounted by the proportion of nonpermuted records. 4 would then force data releasers to focus both on distances but also on the number of records permuted, which is an alternative way to gauge disclosure risk in the permutation paradigm.…”

Section: Definitionmentioning

confidence: 99%

See 4 more Smart Citations

On some consequences of the permutation paradigm for data anonymization: Centrality of permutation matrices, universal measures of disclosure risk and information loss, evaluation by dominance

Ruiz

2018

Information Sciences

View full text Add to dashboard Cite

Recently, the permutation paradigm has been proposed in data anonymization to describe any micro data masking method as permutation, paving the way for performing meaningful analytical comparisons of methods, something that is difficult currently in statistical disclosure control research. This paper explores some consequences of this paradigm by establishing some class of universal measures of disclosure risk and information loss that can be used for the evaluation and comparison of any method, under any parametrization and independently of the characteristics of the data to be anonymized. These measures lead to the introduction in data anonymization of the concepts of dominance in disclosure risk and information loss, which formalise the fact that different parties involved in micro data transaction can all have different sensitivities to privacy and information.

show abstract

Section: A New Roadmap For Data Anonymizationmentioning

confidence: 99%

Section: A New Roadmap For Data Anonymizationmentioning

confidence: 99%

Section: A Class Of Universal Measures Of Disclosure Risk Based On Pementioning

confidence: 99%

Section: A Class Of Universal Measures Of Disclosure Risk Based On Pementioning

confidence: 99%

Section: Definitionmentioning

confidence: 99%

See 3 more Smart Citations

On some consequences of the permutation paradigm for data anonymization: Centrality of permutation matrices, universal measures of disclosure risk and information loss, evaluation by dominance

Ruiz

2018

Information Sciences

View full text Add to dashboard Cite

show abstract

PriTxt: A privacy risk assessment method for text data based on semantic correlation learning

Xiong

Lin

Zhu

et al. 2021

Concurrency and Computation

View full text Add to dashboard Cite

Summary Privacy risk assessment plays a fundamental role in privacy preservation, as it determines the extent to which subsequent processing (such as generalization and obfuscation), should be applied to the sensitive data. However, most existing works on privacy risk assessment have focused on structured data, while unstructured text data remain relatively underexplored due to the complexity of natural language. In this article, we propose a novel method, PriTxt, for evaluating the privacy risk associated with text data by exploiting the semantic correlation. Using definitions derived from the General Data Protection Regulation (GDPR), a de facto standard of privacy preservation in practice, PriTxt first defines the private features that related to individual privacy in order to locate the sensitive words. By using the word2vec algorithm, a word‐embedding model is further constructed to identify the quasi‐sensitive words that are semantically correlated to the private features. The privacy risk of a given text is finally evaluated by aggregating the weighted risks of the sensitive and the quasi‐sensitive words in the text. Experiments on real‐world datasets demonstrate that the proposed PriTxt is effective for conducting risk assessment on text data and further outperforms the traditional methods.

show abstract

$$ P_{\alpha ,\beta } $$-Privacy: A Composable Formulation of Privacy Guarantees for Data Publishing Based on Permutation

Ruiz

2020

Privacy in Statistical Databases

View full text Add to dashboard Cite

Disclosure risk assessment via record linkage by a maximum-knowledge attacker

Cited by 27 publications

References 5 publications

On some consequences of the permutation paradigm for data anonymization: Centrality of permutation matrices, universal measures of disclosure risk and information loss, evaluation by dominance

On some consequences of the permutation paradigm for data anonymization: Centrality of permutation matrices, universal measures of disclosure risk and information loss, evaluation by dominance

PriTxt: A privacy risk assessment method for text data based on semantic correlation learning

$$ P_{\alpha ,\beta } $$-Privacy: A Composable Formulation of Privacy Guarantees for Data Publishing Based on Permutation

Contact Info

Product

Resources

About