De-anonymizing social networks

By Danqi Chen, Botao Hu, Shuo Xie. 2012

719 citations recorded by the source corpus.

How researchers see this paper

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

Citations in the corpus

Faceless Person Recognition: Privacy Implications in Social Media (2016, background): “Privacy of users in social network is a well studied topic in the security community [2, 3, 1, 4].”
LOGAN: Membership Inference Attacks Against Generative Models (2017, background): “mation about the training set. This is a realistic setting, which has been considered extensively in the literature; for instance, social graph knowledge has been used to de-anonymize social networks [42]. Overall, auxiliary/incomplete knowledge of sensitive datasets is a common assumption in litera”
LOGAN: Evaluating Privacy Leakage of Generative Models Using Generative Adversarial Networks (2017, background): “For instance, social graph knowledge has been used to de-anonymize social networks [28].”
Differential Privacy: An Economic Method for Choosing Epsilon (2014, background): “be revealed from studies on the social network structure, much like deanonymization attacks against anonymous Twitter accounts [22].”
Private analysis of graph structure (2011, background): “Attacks on anonymized social network data are studied, for example, in [1, 8].”
Link prediction by de-anonymization: How We Won the Kaggle Social Network Challenge (2011, methodology): “Second, we formulate seed identification—the first step of de-anonymization—as a combinatorial optimization problem, specifically weighted graph matching, in contrast to the pattern search approaches of [6] and [26].”
Technical Privacy Metrics (2015, background): “This metric measures the percentage of incorrectly classified users or events U ′ within the set of all users or events U , for example, users that were incorrectly deanonymized by the adversary [97] or events that were incorrectly classified in a smart metering scenario [83]: privPIC ≡ U ′ U .”
Towards a Visual Privacy Advisor: Understanding and Predicting Privacy Risks in Images (2017, unknown): “Some works [32, 43] study the case where users identity, location or other details can be”
An Analysis of Anonymity in the Bitcoin System (2011, unknown): “Narayanan and Shmatikov [21] deanonymized the Netflix Prize dataset using information from IMDB3 which had similar user content, showing that statistical matching between different but related datasets can be used to attack anonymity.”
Preserving Link Privacy in Social Network Based Systems (2012, background): “Thus in real world settings, X-Vine is vulnerable to the Narayanan-Shmatikov attack [26].”
Anatomy of Threats to the Internet of Things (2019, unknown): “nodes and avoiding de-anonymization attacks [210].”
Privacy wizards for social networking sites (2010, background): “Even if all profile information is removed, it is often possible to re-identify individuals in the published data simply based on unique graph topologies [7, 24, 32].”
Gradient-Leaks: Understanding and Controlling Deanonymization in Federated Learning (2018, background): “This has motivated signiﬁcant research in the area of identifying factors that potentially lead to deanonymization of individuals, such as in social networks [58], programmatic code [2], and product reviews [39].”
Deanonymizing mobility traces: using social network as a side-channel (2012, methodology): “The closest related work is [23] that proposes using graph deanonymization for social networks.”
COSNET: Connecting Heterogeneous Social Networks with Local and Global Consistency (2015, background): “Similar ideas have been studied in [27, 38].”
Toward Reproducibility in Online Social Network Research (2018, background): “ently anonymised datasets simply replace names with random identifiers, rather than obfuscating uniquely identifying attributes, permitting re-identification [26].”
Metrics for Differential Privacy in Concurrent Systems (2014, background): “Differential privacy [14] was originally proposed for privacy protection in the context of statistical databases, but nowadays it is becoming increasingly popular in many other fields, ranging from programming languages [24] to social networks [23] and geolocation [20].”
Plausible Deniability for Privacy-Preserving Data Synthesis (2017, background): “…for example, AOL search log dataset [2], Netﬂix movie rating dataset [39], Genomic data [48, 22], location data [18, 46], and social networks data [40], shows that simple modiﬁcation of sensitive data by removing identiﬁers or by generalizing/suppressing data features results in major information…”
Inferring social ties from geographic coincidences (2010, background): “Other work has shown how social network structure can be exposed by analyzing anonymized versions of it (18, 19) or by looking for commonalities in online behavior, such as covisitations to web sites (20) and tagging shared content with similar textual keywords (21).”