Model-Based Context Privacy for Personal Data Streams [Poster]

NESL Technical Report #: 2012-8-3


Abstract: Smart phones with increased computation and sensing ca- pabilities have enabled the growth of a new generation of applications which are organic and designed to react de- pending on the user contexts. These contexts typically de- fine the personal, social, work and urban spaces of an in- dividual and are derived from the underlying sensor mea- surements. The shared context streams therefore embed in them information, which when stitched together can reveal behavioral patterns and possible sensitive inferences, raising serious privacy concerns. In this paper, we propose a model based technique to capture the relationship between these contexts, and better understand the privacy implications of sharing them. We further demonstrate that by using a gen- erative model of the context streams we can simultaneously meet the utility objectives of the context-aware applications while maintaining individual privacy. We present our cur- rent implementation which uses offline model learning with online inferencing performed on the smart phone. Prelimi- nary results are presented to provide proof-of-concept of our proposed technique.

Publication Forum: 19th ACM Conference on Computer and Communications Security

Date: 2012-08-01

NESL Document?: Yes

Document category: Poster