Browsing by Author "TAN, Kian Lee"

Now showing 1 - 2 of 2
  • No Thumbnail Available
    Item
    CASTLE: Continuously Anonymizing Data Streams
    (2008-02-26) CAO, Jianneng; CARMINATI, Barbara; FERRARI, Elena; TAN, Kian Lee
    Most of existing privacy preserving techniques such as k-anonymity methods, are designed for static data sets. As such, they cannot be applied to streaming data which are continuous, transient and usually unbounded. Moreover, in streaming applications, there is a need to offer strong guarantees on the maximum allowed delay between incoming data and the corresponding anonymized output. To cope with these requirements, in this paper, we present CASTLE (Continuously Anonymizing STreaming data via adaptive cLustEring), a cluster-based scheme that anonymizes data streams on-the-fly and, at the same time, ensures the freshness of the anonymized data by satisfying specified delay constraints. We further show how CASTLE can be easily extended to handle l-diversity. Our extensive performance study shows that CASTLE is efficient and effective wrt the quality of the output data.
  • No Thumbnail Available
    Item
    Publishing Trajectory with Differential Privacy: A Priori vs A Posteriori Sampling Mechanisms
    (2013-04-16T03:02:38Z) SHAO, Dongxu; JIANG, Kaifeng; KISTER, Thomas; BRESSAN, Stephane; TAN, Kian Lee
    It is now possible to collect and share trajectory data for any ship in the world by various means such as satellite and VHF systems. However, the publication of such data also creates new risks for privacy breach with consequences on the security and liability of the stakeholders. Thus, there is an urgent need to develop methods for preserving the privacy of published trajectory data. In this paper, we propose and comparatively investigate two mechanisms for the publication of the trajectory of individual ships under differential privacy guarantees. Traditionally, privacy and differential privacy is achieved by perturbation of the result or the data according to the sensitivity of the query. Our approach, instead, combines sampling and interpolation. We present and compare two techniques in which we sample and interpolate (a priori) and interpolate and sample (a posteriori), respectively. We show that both techniques achieve a $(0, \delta)$ form of differential privacy. We analytically and empirically, with real ship trajectories, study the privacy guarantee and utility of the methods.