TY  - CONF
ID  - www200954
UR  - http://www2009.eprints.org/54/
A1  - Zheleva, Elena
A1  - Getoor, Lise
Y1  - 2009/04//
N2  - In order to address privacy concerns, many social media websites allow users to hide their personal pro?les from the public. In this work, we show how an adversary can exploit an online social network with a mixture of public and private user pro?les to predict the private attributes of users. We map this problem to a relational classi?cation problem and we propose practical models that use friendship and group membership information (which is often not hidden) to infer sensitive attributes. The key novel idea is that in addition to friendship links, groups can be carriers of signi?cant information. We show that on several well-known social media sites, we can easily and accurately recover the information of private-pro?le users. To the best of our knowledge, this is the ?rst work that uses link-based and group-based classi?cation to study privacy implications in social networks with mixed public and private user pro?les.
TI  - To Join or Not to Join: The Illusion of Privacy in Social Networks with Mixed Public and Private User Profiles
SP  - 531
M2  - Madrid, Spain
AV  - public
EP  - 531
T2  - 18th International World Wide Web Conference
ER  -