|
报告人学术简历
|
Chuan Ma
received the B.S. degree from the Beijing University of Posts and
Telecommunications (BUPT) in 2013, and the Doctor of Philosophy (Ph.D.)
degree in Telecommunication from the University of Sydney (USYD) in 2018. He
is now currently working as a lecture at the School of Electrical and Optical
Engineering, Nanjing University of Science and Technology, Nanjing, China. He
has published more than 10 transaction and conference papers, including a
best paper in WCNC 2018. His research interests include stochastic geometry,
device-to-device communication, wireless caching networks and machine
learning, and now working on the big data privacy.
|
|
报告内容
框架
|
There is a
growing trend towards attacks on database privacy due to great value of
privacy information stored in big dataset. In this work, we propose an
stochastic perturbation method to improve the privacy level of the sanitized
data set. Different from most existing works, which calibrate noise to the
dataset, we provide a new influence perturbation algorithm to sanitize the
data record. In this work, we not only prove that the proposed method is
satisfied with the ε-differential privacy, but also derive the expression of
the utility level. Therefore, the tradeoff between the privacy and utility
level is also investigated against different system parameters. Our
simulation results show that, compared with other perturbation methods, the
proposed aggregation algorithm can be a more effective and superior tool to
maintain the privacy level, and achieve a high utility level at the same
time. Moreover, we also investigate that this tradeoff can be adjusted by
changing the value of parameters in the algorithm.
|
