Poster: Constructing a Unique Profile for Mobile User Identification in Location Recommendation Systems

It has been established in previous research that only a small number of spatio-temporal points are enough to uniquely identify an individual [1]. This means, if a user u visited the set of locations {a,b,. . . ,z} then only a small number of these locations would be enough to prove the uniqueness of the mobility traces of u. In this research however, we argue that a profile constructed from such a small set of spatio-temporal points would not be very useful in the context of location prediction and recommendation. Indeed in such context, finding a distinct set of data that makes the individual unique is not the key point. It is much more useful to have a rich profile that, in addition to being unique also reflects the individual’s interest in terms of the places that they visit and the activities that they undertake. Such a profile clearly offers a distinct advantage where it allows grouping together individuals with similar interest and taste. The ability to create such grouping is the foundation upon which collaborative prediction and recommendation systems are developed. Setting aside the sensitive privacy issues, we have been investigating the possibility of constructing a dynamic method of identification using mobility data which, for each individual user possesses measurable variations that make it suitable for ’mobility fingerprinting’.