www.icact.org

TACT Journal Page

Call for Paper

Author Page

Paper Procedure

Paper Submission

Registration

Welcome Message

Statistics

Committee

Paper Archives

Outstanding Papers

Proceedings

Presentation Assistant

Hotel & Travel Info

Photo Gallery

FAQ

Member Login

Scheduler Login

Seminar

Archives Login

Sponsors

Virual Oral Question and Answer


IEEE/ICACT20230212 Question.4 Questioner: nvlinh@cs.ccu.edu.tw 2023-02-17 오전 11:47:32	IEEE/ICACT20230212 Answer.4 Answer by Auhor 48838968@qq.com 2023-02-17 오전 11:47:32
What are your contributions?	With the acceleration of the process of data opening and sharing in the power industry, the risk of sensitive data leakage it faces is also gradually increasing. Privacy protection is one of the hot issues in the research of privacy disclosure control technology in data publishing at present, and k-anonymity is the hotspot of privacy protection research in recent years. In this paper, a hierarchical DP-K anonymous data publishing model based on binary tree clustering is proposed to solve the problem that the existing k-anonymity schemes lack the consideration of attribute sensitivity and minimize the amount of information loss. A clustering algorithm based on binary tree (BTCA) is proposed to divide similar data records into the same equivalent class, which can improve the clustering effect, reduce the information loss caused by the publication of anonymous data sets, and improve the availability of data. The clustered anonymous data sets are allocated different privacy budgets according to the privacy weight of the quasi-identifier attribute, and the hierarchical protection of different sensitive data is realized through the differential privacy noise mechanism, which enhances the privacy of the data.

Select Voice