Efficient Time-Stamped Event Sequence Anonymization

Reza Sherkat, Jing Li, Nikos Mamoulis
2013 ACM Transactions on the Web  
With the rapid growth of applications which generate timestamped sequences (click streams, RFID sequences, GPS trajectories), sequence anonymization becomes an important problem, if such data should be published or shared. Existing trajectory anonymization techniques disregard the importance of time or the sensitivity of events. This paper is the first, to our knowledge, thorough study on timestamped event sequence anonymization. We propose a novel and tunable generalization framework tailored
more » ... o event sequences. We generalize timestamps using time intervals and events using a taxonomy which models the domain semantics. We consider two scenarios: (i) sharing the data with a single receiver; in this case, the receiver's background knowledge is confined to a set of timestamps and time generalization suffices, and (ii) sharing the data with multiple receivers; in this case, time generalization should be combined with event generalization. For both cases, we propose appropriate anonymization methods that prevent both user identification and event prediction. Extensive experiments confirm the efficiency and the scalability of our techniques and demonstrate the quality of the produced anonymizations.
doi:10.1145/2532643 fatcat:qwr7vu3s7bcctarujcik3n3suy