Publishing Community-Preserving Attributed Social Graphs with a Differential Privacy Guarantee

@article{Chen2020PublishingCA,
  title={Publishing Community-Preserving Attributed Social Graphs with a Differential Privacy Guarantee},
  author={Xihui Chen and Sjouke Mauw and Yunior Ram{\'i}rez-Cruz},
  journal={Proceedings on Privacy Enhancing Technologies},
  year={2020},
  volume={2020},
  pages={131 - 152}
}
Abstract We present a novel method for publishing differentially private synthetic attributed graphs. Our method allows, for the first time, to publish synthetic graphs simultaneously preserving structural properties, user attributes and the community structure of the original graph. Our proposal relies on CAGM, a new community-preserving generative model for attributed graphs. We equip CAGM with efficient methods for attributed graph sampling and parameter estimation. For the latter, we… 

Figures and Tables from this paper

MIPAS VALIDATION WITH THE DLR FALCON Andreas Fix

The DLR airborne water vapour differential absorption lidar (H2O DIAL) was employed as the core instrument on board the research aircraft Falcon 20 to validate the water vapour products of MIPAS on

Locally Differentially Private Analysis of Graph Statistics

TLDR
This work considers a local model, and presents algorithms for counting subgraphs -- a fundamental task for analyzing the connection patterns in a graph -- with LDP (Local Differential Privacy), and shows that it is indeed possible to accurately estimate subgraph counts in the local differential privacy model.

Time-Aware Anonymization of Knowledge Graphs

TLDR
The Time-Aware Knowledge Graph Anonymization Algorithm is developed to anonymize KGs such that all published anonymized versions of a KG satisfy the (k, l)-sad principle, by preserving the utility of the anonymized data.

SoK: Differential Privacy on Graph-Structured Data

TLDR
This work systematise different formulations of DP on graphs, discuss challenges and promising applications, including the GNN domain, and compares and separate works into graph analysis tasks and graph learning tasks with GNNs.

Preventing active re-identification attacks on social graphs via sybil subgraph obfuscation

TLDR
This paper shows that k-symmetry, a privacy property introduced in the context of passive attacks, provides a sufficient condition for the protection against active re-identification attacks leveraging an arbitrary number of sybil nodes, and shows that the algorithm K-Match, originally devised for efficiently enforcing the related notion of k-automorphism, also guarantees k-Symmetry.

Communication-Efficient Triangle Counting under Local Differential Privacy

TLDR
This work proposes two-rounds algorithms consisting of edge sampling and carefully selecting edges each user downloads so that the estimation error is small, and proposes a double clipping technique, which clips the number of edges and then theNumber of noisy triangles, to significantly reduce the sensitivity of each user’s query.

Differentially Private Attributed Network Releasing Based on Early Fusion

TLDR
This paper combines the graph topological structure and the attributes together with a private probability model and generates a synthetic network satisfying differential privacy, to meet the request of attributed networks and achieve high data utility.

A User-Centric Mechanism for Sequentially Releasing Graph Datasets under Blowfish Privacy

TLDR
This article introduces a graph model that is augmented with a time dimension and sampled at discrete time steps and shows that the direct application of state-of-the-art privacy-preserving Differential Private techniques is weak against background knowledge attacker models.

Differential Privacy for Evolving Network Based on GHRG

TLDR
GHRG-DP, a novel differentially private evolving networks releasing algorithm which reduces the noise scale and achieves high data utility is proposed and is shown that it can privately release evolving networks with high accuracy.

References

SHOWING 1-10 OF 78 REFERENCES

Adaptive consensus with a virtual leader of multiple agents governed by locally Lipschitz nonlinearity

This paper is concerned with the second‐order consensus problem of multi‐agent systems with a virtual leader, where all agents and the virtual leader share the same intrinsic dynamics with a locally

Pilin-gene phase variation of Moraxella bovis is caused by an inversion of the pilin genes

TLDR
It is reported that DNAs isolated from bacteria expressing alpha pilin have hybridization patterns consistently different from those of bacteria expressing beta pilin, indicating an inversion of about 2 kilobases of DNA.

The common vector approach and its relation to principal component analysis

The main point of the paper is to show the close relation between the nonzero principal components and the difference subspace together with the complementary close relation between the zero

Growth Regressions and What the Textbooks Don't Tell You

The paper discusses three econometric problems that are rarely given adequate discussion in textbooks: model uncertainty, parameter heterogeneity, and outliers. Leamer's extreme bounds analysis can

The impact of channel bonding on 802.11n network management

TLDR
The goal of this work is to understand the characteristics of channel bonding in 802.11n networks and the factors that influence that behavior to ultimately be able to predict behavior so that network performance is maximized.

Detecting Communities under Differential Privacy

TLDR
This paper chooses Louvain method as the back-end community detection for input perturbation schemes and designs ModDivisive using exponential mechanism with the modularity as the score, which steadily gives the best modularity and avg.F1Score on large graphs whileLouvainDP outperforms the remaining input perturbedation competitors in certain settings.

Publishing Attributed Social Graphs with Formal Privacy Guarantees

TLDR
This work adapts existing graph models, and introduces a new one, and shows how to augment them with meaningful privacy, and provides a complete workflow, where the input is a sensitive graph, and the output is a realistic synthetic graph.

Involvement of Carbon Source and Acetyl Phosphate in the External-pH-Dependent Expression of Porin Genes inEscherichia coli

TLDR
The porin composition of the Escherichia coli cell envelope was analyzed during growth at different external pHs as a function of the acetyl phosphate level, leading to modulation of the amount of OmpR-P and subsequent pHo-dependent expression of ompF and ompC.
...