Time Distortion Anonymization for the Publication of Mobility Data - - PowerPoint PPT Presentation

▶
time distortion anonymization for the publication of
SMART_READER_LITE
LIVE PREVIEW

Time Distortion Anonymization for the Publication of Mobility Data - - PowerPoint PPT Presentation

Apr 08, 2023 344 likes •628 views

Time Distortion Anonymization for the Publication of Mobility Data with High Utility Vincent Primault, Sonia Ben Mokhtar, Cdric Lauradoux and Lionel Brunie Mobility data usefulness Real-time traffic, traffic prediction Companies

slide-1
SLIDE 1

Time Distortion Anonymization for the Publication of Mobility Data with High Utility

Vincent Primault, Sonia Ben Mokhtar, Cédric Lauradoux and Lionel Brunie

slide-2
SLIDE 2

Mobility data usefulness…

2

Real-time traffic, traffic prediction Long-term place prediction Companies collecting data

slide-3
SLIDE 3

… and threats

Gambs et al. Show Me How You Move and I Will Tell You Who You Are. Transactions

  • n Data Privacy, 2011.

3

slide-4
SLIDE 4
slide-5
SLIDE 5

Privacy-preserving data publication

5

Anonymized mobility traces Partially re-identified mobility traces Clustering White pages Public maps Raw mobility traces Protection mechanism Attacker Useful analysis results Researcher Data mining Machine learning Simulations

slide-6
SLIDE 6

Outline

  • Introduction
  • State of the art
  • Making a PROMESSE
  • Experimental evaluation
  • Conclusion

6

slide-7
SLIDE 7

A mobility trace

7

A trace is a temporally

  • rdered list of records

belonging to a same user. A record is a triplet (user, location, timestamp).

slide-8
SLIDE 8

Extraction of points of interest (POIs)

8

Done by using, e.g., an appropriate clustering algorithm. Points of interest convey semantic information about habits and can lead to users re-identification.

slide-9
SLIDE 9

Location privacy protection mechanisms for data publication

9

k-anonymity Differential privacy

Wait For Me [Abul et al., 2010] Geo-Indistinguishability [Andrés et al., 2013] Abul et al. Anonymization of moving objects databases by clustering and

  • perturbation. Information Systems, 2010.

Andrés et al. Geo-indistinguishability: Differential privacy for Location-based Systems. CCS, 2013.

slide-10
SLIDE 10

Wait For Me

∂ represents the incertitude that comes from GPS measurements. Wait For Me enforces (k,∂)-anonymity, i.e., there is always at least k users in a cylinder of radius ∂/2.

10

Abul et al. Anonymization of moving objects databases by clustering and

  • perturbation. Information Systems, 2010.

∂ ∂

slide-11
SLIDE 11

Geo-Indistinguishability

11

Andrés et al. Geo-indistinguishability: Differential privacy for Location-based Systems. CCS, 2013. l1, r1 l2, r2 l3, r3 l4, r4 Level of privacy li within ri proportional to an ε

Real location Protected location

slide-12
SLIDE 12

Outline

  • Introduction
  • State of the art
  • Making a PROMESSE
  • Experimental evaluation
  • Conclusion

12

slide-13
SLIDE 13

Intuition behind our work

No state-of-the-art mechanism is both privacy- preserving and useful for data scientists.

13

We believe geographical information is the most important one, so we propose a new mechanism that minimally distort the location. Almost all of them alter the geographical information in some way.

slide-14
SLIDE 14

Hiding POIs with speed smoothing

The idea To guarantee a constant speed along a trace. More challengingto identify where a user stops, and therefore her POIs. How? Divide traces into smaller trajectories, typically one day long. Enforce an equal duration and length between two consecutive records.

14

slide-15
SLIDE 15

Speed smoothing

15

10:05 10h08 10:05 10:08 10:07 10:06 epsilon 10:06 10:07 Point of interest

slide-16
SLIDE 16

Outline

  • Introduction
  • State of the art
  • Making a PROMESSE
  • Experimental evaluation
  • Conclusion

16

slide-17
SLIDE 17

Experimenting with three real-life datasets

Cabspotting Geolife MDC Records 8,9M 3,8M 1,1M Traces 5,5k 2,4k 4,6k Avg trace duration 32 h 3 h 3 h Avg sampling rate 72 s 7 s 32 s

17

slide-18
SLIDE 18

POIs retrieval

POIs with maximum diameter of 200 meters and minimum duration of 15 minutes. Two POIs match if their centroids are within 100 meters.

18

slide-19
SLIDE 19

POIs retrieval (F-score)

19

0% 10% 20% 30% 40% 50% 60% Cabspotting Geolife MDC

Lower is better

slide-20
SLIDE 20

Average spatial error

20

Real trace Protected trace

slide-21
SLIDE 21

Average spatial error

21

0,1 1 10 100 1000 10000 100000 Spatial error, in meters Cabspotting Geolife MDC

Lower is better (log scale)

slide-22
SLIDE 22

Range queries distortion

22

From 2 to 8 hours 1,000 different queries

Distortion is |Q(D) – Q(D’)|/Q(D)

slide-23
SLIDE 23

Range queries distortion

23

0% 20% 40% 60% 80% 100% 120% Cabspotting Geolife MDC

Lower is better

slide-24
SLIDE 24

Outline

  • Introduction
  • State of the art
  • Making a PROMESSE
  • Experimental evaluation
  • Future work
  • Conclusion

24

slide-25
SLIDE 25

Summary

  • Introduced time distortion, opened a new

research direction.

  • Implemented a new protection mechanism for

data publishing, addressing a severe threat while maintaining high utility.

  • Evaluated against three real-life datasets.

25

slide-26
SLIDE 26

Questions

26