Skip to Main Content
 

Global Search Box

 
 
 
 

Files

File List


Social_Network_Data_Privacy .pdf (881.57 KB)

ETD Abstract Container

Abstract Header

Privacy Preserving Social Network Data Publishing

Lin, Zehua
http://orcid.org/0000-0002-0587-0000
http://rave.ohiolink.edu/etdc/view?acc_num=miami1610045108271476

Abstract Details

2021, Master of Computer Science, Miami University, Computer Science and Software Engineering.
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.
Vaskar Raychoudhury (Advisor)
Daniela Inclezan (Committee Member)
Khodakhast Bibak (Committee Member)
43 p.
Computer Science

Recommended Citations

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)

miami1610045108271476
197
© 2021, all rights reserved.
This open access ETD is published by Miami University and OhioLINK.