Frontiers of Data and Computing ›› 2023, Vol. 5 ›› Issue (2): 150-163.

CSTR: 32002.14.jfdc.CN10-1649/TP.2023.02.012

doi: 10.11871/jfdc.issn.2096-742X.2023.02.012

• Technology and Application • Previous Articles     Next Articles

A KSPPL-Anonymity Algorithm for Personalized Location Data Publishing

LU Gongpu(),LI Xiaohui*()   

  1. School of Electronics & Information Engineering, Liaoning University of Technology, Jinzhou, Liaoning 121000, China
  • Received:2022-02-21 Online:2023-04-20 Published:2023-04-24
  • Contact: LI Xiaohui E-mail:lgplwx@163.com;lhxlxh@163.com;lhxlxh@163.com

Abstract:

[Objective] Location data contains a large amount of user privacy information. In location data publishing, if the original data is released directly, it will expose the user's location and other information, which will pose a huge threat to the user's privacy. This phenomenon is even more prominent in continuous location data publishing. Therefore, a personalized location data publishing algorithm KSPPL-Anonymity based on k-anonymity location division is proposed. [Methods] The algorithm improves the efficiency of location k-anonymity by location division. Aiming at the problem that the insertion of noise data will reduce the availability of data, a generation method of noise data is proposed in this algorithm, which improves the availability of data. The disclosure of the user's sensitive location will pose a great threat to the user's privacy, so this algorithm proposes a method to obtain the non-sensitive location with the lowest degree of correlation with the sensitive location, which greatly protects the user's sensitive location from being exposed. By analyzing user location data through time series, the privacy leakage caused by users staying in sensitive locations for a long time and replacing them with the same sensitive locations can be avoided. [Results] Experiments show that, compared with the previous location data publishing methods, this algorithm has some improvements in data availability, privacy protection, and running efficiency. [Conclusions] The algorithm proposed in this paper can better protect users' privacy, meet users' personalized privacy protection needs and ensure the availability of data.

Key words: big data, location data release, k-anonymity, grouping technology, optimal correlation, Lagrange multiplier rule