TY - JOUR
T1 - Lclean
T2 - A Plausible Approach to Individual Trajectory Data Sanitization
AU - Han, Qilong
AU - Lu, Dan
AU - Zhang, Kejia
AU - Du, Xiaojiang
AU - Guizani, Mohsen
N1 - Publisher Copyright:
© 2013 IEEE.
PY - 2018/5/6
Y1 - 2018/5/6
N2 - In recent years, with the continuous development of significant data industrialization, trajectory data have more and more critical analytical value for urban construction and environmental monitoring. However, the trajectory contains a lot of personal privacy, and rashly publishing trajectory data set will cause serious privacy leakage risk. At present, the privacy protection of trajectory data mainly uses the methods of data anonymity and generalization, without considering the background knowledge of attackers and ignores the risk of adjacent location points may leak sensitive location points. In this paper, based on the above problems, combined with the location correlation of trajectory data, we proposed a plausible replacement method. First, the correlation of trajectory points is proposed to classify the individual trajectories containing sensitive points. Then, according to the relevance of location points and the randomized response mechanism, a reasonable candidate set is selected to replace the sensitive points in the trajectory to satisfy the locally differential privacy. Theoretical and experimental results show that the proposed method not only protects the sensitive information of individuals but also does not affect the overall data distribution.
AB - In recent years, with the continuous development of significant data industrialization, trajectory data have more and more critical analytical value for urban construction and environmental monitoring. However, the trajectory contains a lot of personal privacy, and rashly publishing trajectory data set will cause serious privacy leakage risk. At present, the privacy protection of trajectory data mainly uses the methods of data anonymity and generalization, without considering the background knowledge of attackers and ignores the risk of adjacent location points may leak sensitive location points. In this paper, based on the above problems, combined with the location correlation of trajectory data, we proposed a plausible replacement method. First, the correlation of trajectory points is proposed to classify the individual trajectories containing sensitive points. Then, according to the relevance of location points and the randomized response mechanism, a reasonable candidate set is selected to replace the sensitive points in the trajectory to satisfy the locally differential privacy. Theoretical and experimental results show that the proposed method not only protects the sensitive information of individuals but also does not affect the overall data distribution.
KW - Correlation of points
KW - locally differential privacy
KW - sensitive points
KW - trajectory data
UR - http://www.scopus.com/inward/record.url?scp=85046808900&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85046808900&partnerID=8YFLogxK
U2 - 10.1109/ACCESS.2018.2833163
DO - 10.1109/ACCESS.2018.2833163
M3 - Article
AN - SCOPUS:85046808900
VL - 6
SP - 30110
EP - 30116
JO - IEEE Access
JF - IEEE Access
ER -