Skip to Main Content
Frequently Asked Questions
Submit an ETD
Global Search Box
Need Help?
Keyword Search
Participating Institutions
Advanced Search
School Logo
Files
File List
Social_Network_Data_Privacy .pdf (881.57 KB)
ETD Abstract Container
Abstract Header
Privacy Preserving Social Network Data Publishing
Author Info
Lin, Zehua
ORCID® Identifier
http://orcid.org/0000-0002-0587-0000
Permalink:
http://rave.ohiolink.edu/etdc/view?acc_num=miami1610045108271476
Abstract Details
Year and Degree
2021, Master of Computer Science, Miami University, Computer Science and Software Engineering.
Abstract
Publishing data on social networks often violates our privacy. People think that the data they publish is not sensitive, but attackers can get important information from it. While we cannot and should not stop people from publishing data, we do not want people to give away their privacy. Existing approaches while separately consider inference attacks on personal attributes and friendship, none has considered them jointly. This is not fully secure as an attacker can easily discover the target user's privacy attributes through friends. In addition, no previous works have considered corresponding data sanitization methods. In this dissertation, we shall simulate the attackers carrying out inference attacks from the aspects of personal attributes and friendship information embedded in user-generated data. We shall also adopt a two-stage data sanitization method to protect sensitive user attributes. While the first stage sanitizes personal attributes, the second stage sanitizes friendship information. We have found that our first-stage data sanitization renders attacks 20% less effective at predicting user attributes. Our second-stage data sanitization further reduces the prediction accuracy of an attacker by 10%. Extensive experiments show that unlike any previous method, our method of two- stage data sanitization can more effectively protect user privacy.
Committee
Vaskar Raychoudhury (Advisor)
Daniela Inclezan (Committee Member)
Khodakhast Bibak (Committee Member)
Pages
43 p.
Subject Headings
Computer Science
Recommended Citations
Refworks
EndNote
RIS
Mendeley
Citations
Lin, Z. (2021).
Privacy Preserving Social Network Data Publishing
[Master's thesis, Miami University]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=miami1610045108271476
APA Style (7th edition)
Lin, Zehua.
Privacy Preserving Social Network Data Publishing .
2021. Miami University, Master's thesis.
OhioLINK Electronic Theses and Dissertations Center
, http://rave.ohiolink.edu/etdc/view?acc_num=miami1610045108271476.
MLA Style (8th edition)
Lin, Zehua. "Privacy Preserving Social Network Data Publishing ." Master's thesis, Miami University, 2021. http://rave.ohiolink.edu/etdc/view?acc_num=miami1610045108271476
Chicago Manual of Style (17th edition)
Abstract Footer
Document number:
miami1610045108271476
Download Count:
197
Copyright Info
© 2021, all rights reserved.
This open access ETD is published by Miami University and OhioLINK.