A dummy location generation algorithm based on the semantic quantification of location
Xiujin Shi, Junrui Zhang, Yuan Gong · 2021
With the widespread use of location-based services, the risk of location privacy leakage increases. Aiming at the problem that the traditional privacy protection scheme does not fully consider the location privacy leakage caused by the attacker's ability to infer the attack through location semantics, this paper proposes a dummy location generation algorithm based on semantic quantification of location (Virtual Location -based on Semantics, VLBS). The location semantics were quantified by the number of users visiting in different time periods, and a multi-objective optimized dummy location set was constructed from three aspects: historical query probability, semantics of location and physical dispersion uniformity. Experiments show that, compared with other algorithms, the proposed scheme improves the anonymity success rate by 45.6%, the anonymity time by 20.5%, and the physical dispersion uniformity by 12.7%.