Practical Estimation of Mutual Information on Non-Euclidean Spaces - International Cross Domain Conference for Machine Learning and Knowledge Extraction (CD-MAKE 2017)
Conference Papers Year : 2017

Practical Estimation of Mutual Information on Non-Euclidean Spaces

Abstract

We propose, in this paper, to address the issue of measuring the impact of privacy and anonymization techniques, by measuring the data loss between “before” and “after”. The proposed approach focuses therefore on data usability, more than in ensuring that the data is sufficiently anonymized. We use Mutual Information as the measure criterion for this approach, and detail how we propose to measure Mutual Information over non-Euclidean data, in practice, using two possible existing estimators. We test this approach using toy data to illustrate the effects of some well known anonymization techniques on the proposed measure.
Fichier principal
Vignette du fichier
456304_1_En_9_Chapter.pdf (348.59 Ko) Télécharger le fichier
Origin Files produced by the author(s)
Loading...

Dates and versions

hal-01677135 , version 1 (08-01-2018)

Licence

Identifiers

Cite

Yoan Miche, Ian Oliver, Wei Ren, Silke Holtmanns, Anton Akusok, et al.. Practical Estimation of Mutual Information on Non-Euclidean Spaces. 1st International Cross-Domain Conference for Machine Learning and Knowledge Extraction (CD-MAKE), Aug 2017, Reggio, Italy. pp.123-136, ⟨10.1007/978-3-319-66808-6_9⟩. ⟨hal-01677135⟩
236 View
120 Download

Altmetric

Share

More