Correcting Finite Sampling Issues in Entropy l-diversity

Stammler, Sebastian; Katzenbeisser, Stefan; Hamacher, Kay

doi:10.1007/978-3-319-45381-1_11

Cited by 4 publications

(3 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The empirical counts n i are used to obtain maximum-likelihood estimators of the respective pi = n i /L, where L = ∑ i n i is the number of all observations. Note that for "small" L, we would need to include corrections to the Shannon entropy formula to account for small coverage [12][13][14] for which efficient computational methods exist [15]. In this study, we deal with a synthetic dataset and two real-life ones that are sufficiently large so that we can avoid this complication here.…”

Section: Methodsmentioning

confidence: 99%

Consistent Quantification of Complex Dynamics via a Novel Statistical Complexity Measure

Keul

Hamacher

2022

Entropy

Self Cite

View full text Add to dashboard Cite

Natural systems often show complex dynamics. The quantification of such complex dynamics is an important step in, e.g., characterization and classification of different systems or to investigate the effect of an external perturbation on the dynamics. Promising routes were followed in the past using concepts based on (Shannon’s) entropy. Here, we propose a new, conceptually sound measure that can be pragmatically computed, in contrast to pure theoretical concepts based on, e.g., Kolmogorov complexity. We illustrate the applicability using a toy example with a control parameter and go on to the molecular evolution of the HIV1 protease for which drug treatment can be regarded as an external perturbation that changes the complexity of its molecular evolutionary dynamics. In fact, our method identifies exactly those residues which are known to bind the drug molecules by their noticeable signal. We furthermore apply our method in a completely different domain, namely foreign exchange rates, and find convincing results as well.

show abstract

Section: Methodsmentioning

confidence: 99%

Consistent Quantification of Complex Dynamics via a Novel Statistical Complexity Measure

Keul

Hamacher

2022

Entropy

Self Cite

View full text Add to dashboard Cite

show abstract

“…The development of new data anonymization methods is another area in which ARX is frequently utilized. An interesting example is the work by Stammler et al, who have used ARX to implement and evaluate an enhanced variant of the ℓ‐diversity privacy model which uses an asymptotically unbiased estimator for the Shannon entropy . Li et al have proposed and implemented a graph‐based framework for privacy‐preserving data publishing, which they evaluated by comparing the output of their framework with the output of ARX .…”

Section: Summary and Practical Experiencesmentioning

confidence: 99%

Flexible data anonymization using ARX—Current status and challenges ahead

et al. 2020

View full text Add to dashboard Cite

The race for innovation has turned into a race for data. Rapid developments of new technologies, especially in the field of artificial intelligence, are accompanied by new ways of accessing, integrating, and analyzing sensitive personal data. Examples include financial transactions, social network activities, location traces, and medical records. As a consequence, adequate and careful privacy management has become a significant challenge. New data protection regulations, for example in the EU and China, are direct responses to these developments. Data anonymization is an important building block of data protection concepts, as it allows to reduce privacy risks by altering data. The development of anonymization tools involves significant challenges, however. For instance, the effectiveness of different anonymization techniques depends on context, and thus tools need to support a large set of methods to ensure that the usefulness of data is not overly affected by risk-reducing transformations. In spite of these requirements, existing solutions typically only support a small set of methods. In this work, we describe how we have extended an open source data anonymization tool to support almost arbitrary combinations of a wide range of techniques in a scalable manner. We then review the spectrum of methods supported and discuss their compatibility within the novel framework. The results of an extensive experimental comparison show that our approach outperforms related solutions in terms of scalability and output data quality-while supporting a much broader range of techniques. Finally, we discuss practical experiences with ARX and present remaining issues and challenges ahead.

show abstract

“…Other approaches use k-anonymity [58], l-diversity [18,56], or differential privacy [42,51] to securely perform GWAS on distributed datasets. Nevertheless, several attacks were proposed shortly after: [25,59,61].…”

Section: Motivationmentioning

confidence: 99%

Large-Scale Privacy-Preserving Statistical Computations for Distributed Genome-Wide Association Studies

Tkachenko

Weinert

Schneider

et al. 2018

Proceedings of the 2018 on Asia Conference on Computer and Communications Security

Self Cite

View full text Add to dashboard Cite

We present privacy-preserving solutions for Genome-Wide Association Studies (GWAS) based on Secure Multi-Party Computation (SMPC). Using SMPC, we protect the privacy of patients when medical institutes collaborate for computing statistics on genomic data in a distributed fashion. Previous solutions for this task lack efficiency and/or use inadequate algorithms that are of limited practical value. Concretely, we optimize and implement multiple algorithms for the χ 2-, G-, and P-test in the ABY framework (Demmler et al., NDSS'15) and evaluate them in a distributed GWAS scenario. Statistical tests generally require advanced mathematical operations. For operations that cannot be calculated in integer arithmetic, we make use of the existing IEEE 754 floating point arithmetic implementation in ABY (Demmler et al., CCS'15). To improve performance, we extend the mixed-protocol capabilities of ABY by optimizing and implementing the integer to floating point conversion protocols of Aliasgari et al. (NDSS'13), which may be of independent interest. Furthermore, we consider extended contingency tables for the χ 2-and G-test that use codeword counts instead of counts for only two alleles, thereby allowing for advanced, realistic analyses. Finally, we consider an outsourcing scenario where two non-colluding semi-trusted third parties process secret-shared input data from multiple institutes. Our extensive evaluation shows, compared to the prior art of Constable et al. (BMC Medical Informatics and Decision Making'15), an improved run-time efficiency of the χ 2-test by up to factor 37x. We additionally demonstrate practicality in scenarios with millions of participants and hundreds of collaborating institutes. CCS CONCEPTS • Security and privacy → Privacy-preserving protocols; Management and querying of encrypted data; Privacy protections;

show abstract

Correcting Finite Sampling Issues in Entropy l-diversity

Cited by 4 publications

References 22 publications

Consistent Quantification of Complex Dynamics via a Novel Statistical Complexity Measure

Consistent Quantification of Complex Dynamics via a Novel Statistical Complexity Measure

Flexible data anonymization using ARX—Current status and challenges ahead

Large-Scale Privacy-Preserving Statistical Computations for Distributed Genome-Wide Association Studies

Contact Info

Product

Resources

About