Privacy and utility preserving sensor-data transformations
File(s)
Accepted version
Author(s)
Malekzadeh, Mohammad
Clegg, Richard G
Cavallaro, Andrea
Haddadi, Hamed
Type
Journal Article
Abstract
Sensitive inferences and user re-identification are major threats to privacy
when raw sensor data from wearable or portable devices are shared with
cloud-assisted applications. To mitigate these threats, we propose mechanisms
to transform sensor data before sharing them with applications running on
users' devices. These transformations aim at eliminating patterns that can be
used for user re-identification or for inferring potentially sensitive
activities, while introducing a minor utility loss for the target application
(or task). We show that, on gesture and activity recognition tasks, we can
prevent inference of potentially sensitive activities while keeping the
reduction in recognition accuracy of non-sensitive activities to less than 5
percentage points. We also show that we can reduce the accuracy of user
re-identification and of the potential inference of gender to the level of a
random guess, while keeping the accuracy of activity recognition comparable to
that obtained on the original data.
when raw sensor data from wearable or portable devices are shared with
cloud-assisted applications. To mitigate these threats, we propose mechanisms
to transform sensor data before sharing them with applications running on
users' devices. These transformations aim at eliminating patterns that can be
used for user re-identification or for inferring potentially sensitive
activities, while introducing a minor utility loss for the target application
(or task). We show that, on gesture and activity recognition tasks, we can
prevent inference of potentially sensitive activities while keeping the
reduction in recognition accuracy of non-sensitive activities to less than 5
percentage points. We also show that we can reduce the accuracy of user
re-identification and of the potential inference of gender to the level of a
random guess, while keeping the accuracy of activity recognition comparable to
that obtained on the original data.
Date Issued
2020-03
Date Acceptance
2019-11-07
Citation
Pervasive and Mobile Computing, 2020, 63, pp.1-13
ISSN
1574-1192
Publisher
Elsevier
Start Page
1
End Page
13
Journal / Book Title
Pervasive and Mobile Computing
Volume
63
Copyright Statement
© 2020 Elsevier Ltd. All rights reserved. This manuscript is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International Licence http://creativecommons.org/licenses/by-nc-nd/4.0/
Sponsor
Engineering & Physical Science Research Council (EPSRC)
Engineering & Physical Science Research Council (E
Identifier
http://arxiv.org/abs/1911.05996v1
Grant Number
EP/N028260/2
RGS128099 (EP/R03351X/1)
Subjects
cs.LG
cs.LG
cs.HC
eess.SP
stat.ML
Notes
Accepted to appear in Pervasive and Mobile computing (PMC) Journal, Elsevier
Publication Status
Published
Date Publish Online
2020-03-02