Imperial College London
Portrait Picture

ProfessorMichaelBronstein

Faculty of Engineering, Department of Computing

Visiting Professor
 
 
 
//

Contact

 

m.bronstein Website

 
 
//

Location

 

569Huxley BuildingSouth Kensington Campus

//

Summary

 

Publications

Citation

BibTex format

@article{Cretu:2022:10.1038/s41467-021-27714-6,
author = {Cretu, A-M and Monti, F and Maronne, S and Dong, X and Bronstein, M and de, Montjoye Y},
doi = {10.1038/s41467-021-27714-6},
journal = {Nature Communications},
pages = {1--11},
title = {Interaction data are identifiable even across long periods of time},
url = {http://dx.doi.org/10.1038/s41467-021-27714-6},
volume = {13},
year = {2022}
}
Download

RIS format (EndNote, RefMan)

TY  - JOUR
AB  - Fine-grained records of people’s interactions, both offline and online, arecollected at large scale. These data contain sensitive information about whom wemeet, talk to, and when. We demonstrate here how people’s interaction behavioris stable over long periods of time and can be used to identify individuals inanonymous datasets. Our attack learns the profile of an individual using geometric deep learning and triplet loss optimization. In a mobile phone metadatadataset of more than 40k people, it correctly identifies 52% of individuals basedon their 2-hop interaction graph. We further show that the profiles learned byour method are stable over time and that 24% of people are still identifiableafter 20 weeks. Our results suggest that people with well-balanced interactiongraphs are more identifiable. Applying our attack to Bluetooth close-proximitynetworks, we show that even 1-hop interaction graphs are enough to identifypeople more than 26% of the time. Our results provide strong evidence thatdisconnected and even re-pseudonymized interaction data can be linked together making them personal data under the European Union’s General DataProtection Regulation.
AU  - Cretu,A-M
AU  - Monti,F
AU  - Maronne,S
AU  - Dong,X
AU  - Bronstein,M
AU  - de,Montjoye Y
DO  - 10.1038/s41467-021-27714-6
EP  - 11
PY  - 2022///
SN  - 2041-1723
SP  - 1
TI  - Interaction data are identifiable even across long periods of time
T2  - Nature Communications
UR  - http://dx.doi.org/10.1038/s41467-021-27714-6
UR  - https://www.nature.com/articles/s41467-021-27714-6
UR  - http://hdl.handle.net/10044/1/93168
VL  - 13
ER  -
Download