Towards Trajectory Anonymization: a Generalization-Based Approach

Trajectory datasets are becoming popular due to the massive usage of GPS and location-based services. In this paper, we address privacy issues regarding the identification of individuals in static trajectory datasets. We first adopt the notion of k-anonymity to trajectories and propose a novel generalization-based approach for anonymization of trajectories. We further show that releasing anonymized trajectories may still have some privacy leaks. Therefore we propose a randomization based reconstruction algorithm for releasing anonymized trajectory data and also present how the underlying techniques can be adapted to other anonymity standards. The experimental results on real and synthetic trajectory datasets show the effectiveness of the proposed techniques.