Sensor data shared for receiving personalized services from mobile applications,
embed in them information, which when stitched together can reveal behavioral
patterns and possible sensitive inferences, raising serious privacy concerns.
In this talk, we will present the information theoretic metrics for 
privacy and utility and ways of finding the desired level of data obfuscation
to satisfy the defined metrics. We will then describe the design and
implementation of a context-aware privacy preserving framework on a mobile
platform. Specifically, we will focus on the android-based platform changes. The
framework learns a behavioral model of the user and uses it to
derive the obfuscation function and the extent to which it should be
applied to satisfy the privacy and utility metrics.