Privacy in geo-social context aware systems

With the extensive use of the location sensing technologies in mobile devices, location based applications and services are predicted to be popular. Further, it raises privacy concerns when several user context data alongwith location and time is shared. There is implicit tradeoff between effective usage of Location Based Service that users wish to receive and the location privacy they can afford to risk. Moreover, different users may have varying privacy needs in different contexts. Therefore, it is important to develop customizable or personalized privacy protection mechanisms.

A personalized anonymization model needs to provide user-defined privacy rules to meet user's personal privacy preferences. The location obfuscation techniques that are based on the use of geometric methods also needs to consider location semantics to prevent privacy leaks.Such a obfuscation model needs to be extensible towards a probabilistic model to account varying sensitivity. We introduce a combined approach to achieve more comprehensive solution for privacy preservation in such systems.