Wherefore Art Thou R3579x? Anonymized Social Networks, Hidden Patterns, and Structural Steganography

By Lars Backstrom, Cynthia Dwork, Jon Kleinberg

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

1. Philip S. Yu
2. S. Ji
3. Charu C. Aggarwal
4. Jie Tang
5. Zhilin Yang
6. Chao Li
7. Jian Pei
8. Weiqing Li

Citations in the corpus

• Differentially Private Federated Learning: A Client Level Perspective (2017, background): “Unfortunately, anonymization alone often is not sufficient [7, 2] and standard machine learning approaches largely disregard privacy aspects.”
• Universally utility-maximizing privacy mechanisms (2008, background): “Unfortunately, publication schemes initially thought to be “private” have succumbed to privacy attacks [2, 20, 23], highlighting the urgent need for mechanisms that are provably private.”
• COSNET: Connecting Heterogeneous Social Networks with Local and Global Consistency (2015, background): “[3] presented the processes where one can identify individuals in anonymized networks by either manipulating networks before they are anonymized or by having prior knowledge about certain anonymized nodes.”
• On Your Social Network De-anonymizablity: Quantification and Large Scale Evaluation with Seed Knowledge (2015, methodology): “, [1][2][3]) consist of two phases: seed identification phase which identifies some seed mapping information from V a to V u and mapping propagation phase which propagates the seed mapping information to de-anonymize the rest of the anonymized users.”
• Security issues in cloud environments: a survey (2014, background): “[23] proposed a family of attacks—the active, the passive, and the semi-passive attacks—which breaches edge privacy of a targeted group of individuals on a social network with basis on structural knowledge.”
• Resisting structural re-identification in anonymized social networks (2008, background): “It is not well-understood how publishing a network threatens privacy; initial investigations, including anonymization algorithms, are just emerging [4, 14, 17, 24, 26, 29, 30, 31].”
• Combating Attacks and Abuse in Large Online Communities (2016, background): “Various techniques have been prop osed to compromise user anonymity and infer users’ sensitive information from social network data [158, 152, 253, 43].”
• Structural Data De-anonymization: Quantification, Practice, and Implications (2014, methodology): “However, since the attacks in [1] leverage the success of a “sybil” attack before actual anonymized data publication, they are less practical.”
• To join or not to join: the illusion of privacy in social networks with mixed public and private user profiles (2009, unknown): “If one anonymizes the data näıvely by removing personal attributes and replacing them with a random identifier, it still is possible to identify individuals based on their subgraph structure [3, 8, 14].”
• Predict Anchor Links across Social Networks via an Embedding Approach (2016, background): “However, information about anchor links is often not available in practical scenarios, because most users have no motivation or are un-willing to explicitly correlate their identities in different on-line social networks [Backstrom et al. , 2007].”