Pith. sign in

REVIEW 3 major objections 4 minor 109 references

Practical and Accurate Local Edge Differentially Private Graph Algorithms

T0 review · 3 major / 4 minor · reviewed 2026-08-06 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read Local edge differential privacy can support k-core numbers and triangle counts with error scaling with maximum degree and degeneracy, and the algorithms are the first LEDP graph methods evaluated in a simulated distributed setting on…

desk verdict A genuinely useful algorithm and evaluation, but the triangle-counting privacy proof doesn't add up under the paper's own definition—repairable, but load-bearing. read the letter →

arxiv 2506.20828 v3 pith:N3XUFSRM submitted 2025-06-25 cs.DS cs.CRcs.DB

Pranay Mundra , Charalampos Papamanthou , Julian Shun , Quanquan C. Liu This is my paper · ORCID
classification cs.DScs.CRcs.DB
keywords localedgedifferentialprivacyk-coredecompositiontrianglecountingrandomizedresponsegeometricmechanismdegeneracylowout-degreeorderingdistributedgraphalgorithms
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper aims to establish that two fundamental graph statistics, $k$-core numbers and triangle counts, can be published under local edge differential privacy (LEDP) with error that depends on local structure---maximum degree for $k$-cores and degeneracy for triangles---rather than on the total number of edges. The first algorithm, $\textsc{k-CoreD}$, uses a level data structure with degree-based thresholding and bias terms to return $(2+\eta, O(\log(D_{\max}) \log^2 n / \varepsilon))$-approximate core numbers with high probability, in $O(\log n \log D_{\max})$ rounds. The second, $\textsc{EdgeOrient}\Delta$, orients edges according to that low out-degree ordering, perturbs adjacency lists with randomized response, and returns a $(1+\eta, O(\sqrt{n d} \log^3 n / \varepsilon^2 + \sqrt{\overrightarrow{C_4}}))$-approximation of the triangle count. The paper also reports the first evaluation of LEDP graph algorithms in a simulated distributed setting, with implementations that run on billion-edge graphs and reduce approximation error by orders of magnitude relative to randomized-response baselines.

What carries the argument

The load-bearing mechanism is a level data structure in which each node's final level is produced by noisy move-up checks, with the total number of levels capped by the node's privately noised degree. Degree thresholding is what converts the error dependence from $O(\log^3 n / \varepsilon)$ to $O(\log(D_{\max}) \log^2 n / \varepsilon)$, because thresholds scale with $\log D_{\max}$ instead of $\log n$; the bias terms keep one-sided geometric noise from stranding low-degree nodes at level zero. For triangle counting, the crucial object is the approximate low out-degree ordering produced by $\textsc{k-CoreD}$, where degeneracy $d$ (the largest $k$ such that the graph has a nonempty $k$-core) bounds every node's out-degree; a triangle is then counted by the unique node whose two incident triangle edges are outgoing, and the variance analysis is carried by the number of oriented 4-cycles, bounded by $\widetilde{O}(n^2 d^2)$.

What would settle it

Run the triangle algorithm's four component procedures on an edge-neighboring pair of graphs and compute the worst-case log-likelihood ratio of the full transcript for the edge that differs; if the ratio exceeds $e^{\varepsilon}$ for $\varepsilon = 1$, the per-edge composition in Lemma 4.2 does not hold. A second check is to measure the maximum out-degree produced by $\textsc{k-CoreD}$ on a graph with a dense core: any node whose out-degree exceeds $(2+\eta)d + O(\log(D_{\max}) \log^2 n / \varepsilon)$ would falsify the invariant behind both approximation guarantees.

Watch

Extended reading notes

Core claim

The central claim is that input-dependent private graph properties can replace the edge count as the driver of error in LEDP graph statistics. $\textsc{k-CoreD}$ achieves this by having each node draw symmetric geometric noise once to form a noisy degree, derive a maximum level threshold from that noisy degree, and then climb a level data structure only while its noisy count of same-level neighbors exceeds the current level threshold; two bias terms counter the one-sided effects of geometric noise, and the noise added at each level is scaled by the node's threshold. $\textsc{EdgeOrient}\Delta$ inherits the resulting low out-degree ordering, publishes a randomized-response version of every adjacency list, releases a noisy maximum out-degree, and counts triangles through outgoing edges with Laplace noise scaled by that maximum. The paper proves a $(2+\eta, O(\log(D_{\max}) \log^2 n / \varepsilon))$-approximation for core numbers and a $(1+\eta, O(\sqrt{n d} \log^3 n / \varepsilon^2 + \sqrt{\overrightarrow{C_4}}))$-approximation for triangle counts, both with high probability, where $d$ is the degeneracy and $\overrightarrow{C_4}$ counts oriented 4-cycles. It further reports empirical average $k$-core approximation factors near 3, triangle relative errors near $10^{-2}$, and successful runs on graphs with over a billion edges.

Load-bearing premise

The triangle-counting privacy proof assumes that for each differing edge the combined privacy loss of both endpoints is at most the sum of the four component $\varepsilon/4$ budgets rather than a larger sum that counts both endpoints' triangle-count randomizers separately; if that per-edge accounting fails, the algorithm is not proven $\varepsilon$-LEDP.

Editorial extensions

If this is right

  • On graphs where the maximum degree is far smaller than the number of vertices, $k$-core estimates inherit far less noise than the earlier $\log^3(n)/\varepsilon$ bound, and the algorithm needs only $O(\log n \log D_{\max})$ communication rounds.
  • The low out-degree ordering reduces triangle-count error from terms like $n^{3/2}/\varepsilon^2$ to terms like $\sqrt{n d} \log^3 n / \varepsilon^2$, an improvement of roughly $\sqrt{n}$ on bounded-degeneracy graphs.
  • Because every node perturbs its own adjacency list before any central party sees it, both algorithms fit decentralized and federated deployments where no trusted curator exists.
  • The distributed-simulation experiments indicate that these LEDP algorithms complete within practical time on graphs with over a billion edges, whereas the randomized-response baselines run out of memory or time out on large graphs.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Extension: the same recipe---a private low out-degree ordering followed by local counts restricted to outgoing neighbors---should transfer to other local subgraph patterns (wedges, small cliques, stars), with variance controlled by the number of oriented copies of a slightly larger pattern, just as triangles are controlled by oriented 4-cycles.
  • Extension: since the $k$-core bound is driven by $\log D_{\max}$, a graph with a handful of enormous hubs but small degeneracy could still pay a large noise price; testing a variant that truncates or per-node clips the threshold would reveal whether the bound is tight for such skewed graphs.
  • Extension: a formal per-edge privacy-accounting lemma---tracking each differing edge's total privacy loss across both endpoints---would make the triangle algorithm's composition argument fully explicit and clarify how the budget split ports to other LEDP definitions.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 4 minor

Summary. The manuscript introduces two local edge differentially private (LEDP) graph algorithms: k-CoreD, a k-core decomposition algorithm whose utility bounds depend on the maximum degree, and EdgeOrientΔ, a triangle-counting algorithm that uses a private low out-degree ordering plus randomized response. The paper claims that k-CoreD returns (2+η, O(log(D_max) log^2 n / ε))-approximate core numbers in O(log n log D_max) rounds, and that EdgeOrientΔ returns a (1+η, O(√(n d) log^3 n / ε^2 + √(C4^→)))-approximation of the true triangle count. It also reports the first evaluation of LEDP graph algorithms in a simulated distributed environment, with experiments on graphs up to billion-edge scale for k-core and up to dblp-scale for triangle counting.

Significance. If the proofs are repaired, the paper would be a substantial advance: it replaces purely randomized-response LEDP graph algorithms with input-dependent private mechanisms, improves the theoretical dependence from the number of edges to the maximum degree or degeneracy, and provides a reusable distributed simulation framework with an open-source artifact. The empirical accuracy gains over prior LEDP implementations are large and, for k-core, the method is demonstrated on billion-edge graphs. However, the central privacy proof for triangle counting has a composition gap under the paper's own Definition 2.4, and the k-core analysis does not match the pseudocode's threshold computation. These issues must be resolved before the theoretical guarantees can be accepted.

major comments (3)
  1. [§4.2, Lemma 4.2 and Definition 2.4] The privacy proof for EdgeOrientΔ is not valid under the paper's Definition 2.4. For a fixed edge {u,v}, the sum of the nominal privacy parameters of the local randomizers called by both endpoints is ε/4 (from k-CoreD) plus 2·(ε/4) for the RR round, 2·(ε/4) for the max-out-degree round, and 2·(ε/4) for the triangle-count round, totaling 7ε/4, not ε. The statement that a differing edge 'affects at most one node's out-degree' is a per-edge sensitivity observation, but Definition 2.4 does not allow excluding the other endpoint's local randomizers from the sum. The proof therefore does not establish that EdgeOrientΔ is ε-LEDP. A correct argument requires either a formal per-edge privacy-loss accounting that justifies dropping unaffected randomizers, or a revision of Definition 2.4 and the composition argument.
  2. [§3.1–3.2, Algorithm 3.2 vs. Lemma 3.4] The threshold computation in Algorithm 3.2 uses ⌈log_2(e_d_v)⌉·L, while Lemma 3.4's proof assumes the threshold is computed with log_{1+η/5}(e_d_u) and equates the resulting level with (1+η/5)^{⌊r/(2 log n)⌋}. With the stated L = ⌈log n⌉/4 and the paper's value η = 3.625, these expressions are not compatible: the level reached from a base-2 threshold is not the level implied by the invariant's base-(1+η/5) degree bound. Since the approximation guarantee of Theorem 3.6 rests on Invariants 1 and 2, the analysis and pseudocode must be reconciled, or a corrected derivation that accommodates the actual base-2 threshold must be supplied.
  3. [§4.2, Algorithm 4.2 and §5.2] The claim that the LEDP implementations 'scale to billion-edge graphs' is not supported for EdgeOrientΔ. Section 4.2 states that the RR phase publishes noisy edges for the entire upper-triangular matrix, requiring O(n^2) communication and memory, and the triangle-counting experiments in Figure 10 stop at dblp (≈317k vertices, ≈1M edges), not at billion-edge graphs. The billion-edge experiments are shown only for k-CoreD. The paper should either qualify the scalability claim to k-CoreD or explain how the quadratic RR phase is avoided at billion-edge scale.
minor comments (4)
  1. [§2, Definition 2.9] The density given for Lap(b) is written as 2b·exp(−|X|·b), which is not a probability density as stated; it should be (b/2)·exp(−|X|·b) (or the parameter should be defined as a scale). This makes the Laplace noise scale used in Algorithm 4.4 harder to verify.
  2. [§5, Parameters] The split fraction f = 0.8 and bias term b = 8 are chosen on the evaluation datasets. Since these parameters do not affect the privacy guarantee, this is not a privacy flaw, but the paper should state whether they were tuned on the test graphs and, if so, acknowledge that the reported accuracy numbers reflect that tuning.
  3. [§4.2 and §5.2] The notation e𝑂(·) is used without definition; please define it at first use. Also, the sentence 'Such use cases were unnecessarily in [27,44]' appears to contain a typo and should be rephrased.
  4. [§3.1, Algorithm 3.3] The threshold in Algorithm 3.3 is written as (1+η/5)F(r), which reads as a linear factor, while the invariants and Lemma 3.5 appear to require an exponential threshold (1+η/5)^{F(r)}. Please ensure the superscript formatting is correct and that the pseudocode matches the analysis.

Circularity Check

0 steps flagged · score 0.0 of 10

No circular reduction: k-CoreD and EdgeOrientΔ derive their guarantees from the LEDP definition, prior published baselines, and internal variance/expectation calculations; the apparent Lemma 4.2 privacy composition issue is a correctness gap, not circularity.

full rationale

Walking the derivation chain, no claimed output is equal by construction to an input and no fitted parameter is renamed as a prediction. For k-CoreD, Theorem 3.3 directly instantiates Definition 2.4: degree thresholding is a (ε1/2)-LR, the scaled geometric level-moving noise is (ε2/(2t_v))-LR per round and composes over t_v rounds to ε2/2, and the proof explicitly sums both endpoints' contributions to obtain ε1+ε2=ε. The approximation result imports Invariant 1, Invariant 2, and Theorem 4.1 from Dhulipala et al. (FOCS 2022). Although that paper shares authors with this one, it is a published external result with its own derivation, so this is legitimate support rather than a self-citation loop. EdgeOrientΔ's Lemmas 4.3-4.6 and Theorem 4.7 follow from the k-CoreD ordering guarantee, standard randomized-response debiasing, and a Law-of-Total-Expectation/Variance computation; none of these steps assumes the triangle bound being proved. The experimental choices f=0.8 and b=8 are hyperparameters tuned on the evaluation graphs, which is a generalizability caveat but not a fitted input called a prediction. I flag two non-circular issues: Lemma 4.2's composition may need a per-edge sensitivity argument beyond the literal Definition 2.4 sum over both endpoints' local randomizers, and Lemma 4.3 references a supplementary 'Theorem 3.4' not present in the text; both are correctness/presentation concerns, not circularity, and do not change the score. Thus the paper is self-contained against external published benchmarks and receives 0.

Assumptions & free parameters 3 free parameters · 5 assumptions · 0 invented entities

No new particles, mediators, or synthetic entities are introduced; oriented 4-cycles are a graph statistic used in the analysis, not a postulated entity. The main free parameters are the bias term and the privacy split fraction, both chosen on the evaluation datasets.

free parameters (3)
  • bias term b = 8
    Hand-set constant in Algorithms 3.2 and 3.3; the bias formulas depend on it and experiments use b=8.
  • privacy split fraction f = 0.8
    Selected as 'optimal' from an ablation on gplus, wiki, and livejournal; used for all reported results.
  • approximation constant (2+eta) = 5.625
    Borrowed from non-private k-core experiments [55]; not fitted here but influences the theoretical bounds.
assumptions (5)
  • domain assumption Local edge differential privacy model with honest-but-curious curator and nodes (Definition 2.4)
    The entire privacy guarantee is defined relative to this model; if the model's composition rule is interpreted strictly, the triangle-count proof undercounts budget.
  • standard math Invariants 1 and 2 and Theorem 4.1 from Dhulipala et al. [19]
    The k-core approximation proof invokes these without reproving them; [19] is peer-reviewed but by overlapping authors.
  • domain assumption Bounded degeneracy of real-world graphs (d=O(1))
    Used to convert C4^->=O(n^2) and T=O(n), which makes the triangle error bound sublinear and is claimed for real-world graphs.
  • standard math Unbiasedness of the randomized response estimator after correction
    Standard randomized response correction; used in Lemma 4.5.
  • standard math Laplace and geometric mechanism privacy from standard DP tools
    Background results cited from Dwork et al. and others.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Practical and Accurate Local Edge Differentially Private Graph Algorithms." pith.science (2026). https://pith.science/paper/N3XUFSRM

@misc{pith2026250620828,
  author       = {Pith},
  title        = {Pith review of: Practical and Accurate Local Edge Differentially Private Graph Algorithms},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/N3XUFSRM}},
  note         = {Machine review of arXiv:2506.20828}
}
read the original abstract

The rise of massive networks across diverse domains necessitates sophisticated graph analytics, often involving sensitive data and raising privacy concerns. This paper addresses these challenges using local differential privacy (LDP), which enforces privacy at the individual level, where no third-party entity is trusted, unlike centralized models that assume a trusted curator. We introduce novel LDP algorithms for two fundamental graph statistics: k-core decomposition and triangle counting. Our approach leverages input-dependent private graph properties, specifically the degeneracy and maximum degree of the graph, to improve theoretical utility. Unlike prior methods, our error bounds are determined by the maximum degree rather than the total number of edges, resulting in significantly tighter guarantees. For triangle counting, we improve upon the work of Imola, Murakami, and Chaudhury [USENIX Security `21, `22], which bounds error in terms of edge count. Instead, our algorithm achieves bounds based on graph degeneracy by leveraging a private out-degree orientation, a refined variant of Eden et al.'s randomized response technique [ICALP `23], and a novel analysis, yielding stronger guarantees than prior work. Beyond theoretical gains, we are the first to evaluate local DP algorithms in a distributed simulation, unlike prior work tested on a single processor. Experiments on real-world graphs show substantial accuracy gains: our k-core decomposition achieves errors within 3x of exact values, far outperforming the 131x error in the baseline of Dhulipala et al. [FOCS `22]. Our triangle counting algorithm reduces multiplicative approximation errors by up to six orders of magnitude, while maintaining competitive runtime.

Figures

Figures reproduced from arXiv: 2506.20828 by the authors.

Figure 1
Figure 1. Local edge differential privacy (LEDP) Model [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Example 𝑘-core decomposition and triangles in a 4-degenerate graph. Nodes are assigned core numbers based on the highest value core they belong to; e.g., a node in the 1-core but not in the 2-core is given the core number of 1. Larger valued cores are contained within all smaller valued cores; e.g., the 3-core is contained in the 1 and 2-core. Red edges show the triangles, i.e., 3-cycles in the graph. The degeneracy… view at source ↗
Figure 3
Figure 3. Node movements in 𝑘-CoreD’s Level Data Structure (LDS). Green: active nodes eligible to move; red: thresholded nodes; orange: active nodes that fail the noisy neighbor check. The LDS and threshold structures are shown alongside the graph. Noise is added during the level-moving step to ensure privacy, and snapshots illustrate node progression and halted movement due to thresholding. where 𝑑𝑣 is the true degree and 𝑋 … view at source ↗
Figures from the paper (8 more)
Figure 4
Figure 4. Figure 4: EdgeOrientΔ for blue node: true triangle count is 2, but due to Randomized Response and low out-degree ordering, estimate is 1. 𝜂/5) ⌊𝑟 /(2 log𝑛) ⌋. Thus, using our computed noise, the degree of a node must be at least (1+𝜂/5) ⌊𝑟 /(2 log𝑛) ⌋ −2 log (𝐷max)  2𝑐1 log2 𝑛 …
Figure 5
Figure 5. Figure 5: Oriented cycle of length 4; two non-adjacent black nodes have edges oriented toward the remaining red nodes. 𝑂   𝑑 + log(𝐷max ) log2 (𝑛) 𝜀 2  = 𝑂  𝑑 2 + log2 (𝐷max ) log4 (𝑛) 𝜀 2  outgoing red pairs. Each pair {𝑢, 𝑣 } ⊆ 𝑆𝑤,𝑥 forms an oriented 4-cycle with {𝑤, 𝑥},…
Figure 6
Figure 6. Figure 6: 𝑘-core Decomposition Results [PITH_FULL_IMAGE:figures/full_fig_p013_6.png]
Figure 7
Figure 7. Figure 7: 𝑘-Core Decomposition Avg. Response Time Number of Rounds email-eu-corewiki enron brightkite ego-twitter gplus stanford dblp brain orkut livejournal twitter friendster 102 103 104 k-Core k-CoreD Theoretical Bound (Without Noise) [PITH_FULL_IMAGE:figures/full_fig_p013_7.png]
Figure 8
Figure 8. Figure 8: 𝑘-Core Decomposition Number of Rounds outputs to the coordinator, which aggregates the data and broad￾casts new public information. This proceeds over multiple synchro￾nous rounds, simulating a real-world distributed setting. We use 80 worker processors and a single co…
Figure 9
Figure 9. Figure 9: Avg. approximation factor for 𝑘-Core decomposition vs. epsilon (𝜀) and split fraction (𝑓 ). Time (s) email-eu-core wiki enron brightkite ego-twitter gplus stanford dblp 10−1 101 103 ARROneNS∆ (Lap) EdgeOrient∆ (a) Average Response Time Relative Error email-eu-core wiki…
Figure 10
Figure 10. Figure 10: Triangle Counting Results. ϵ Approximation Factor 0.25 0.50 0.75 1.00 1.50 2.00 100 100.02 100.04 ARROneNS∆ (Lap) EdgeOrient∆ (a) gplus ϵ Approximation Factor 0.25 0.50 0.75 1.00 1.50 2.00 100 102 104 106 ARROneNS∆ (Lap) EdgeOrient∆ (b) stanford ϵ Approximation Factor…
Figure 11
Figure 11. Figure 11: Avg. approx factor for triangle counting vs. epsilon ( [PITH_FULL_IMAGE:figures/full_fig_p014_11.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

109 extracted references · 62 canonical work pages

  1. [1]

    DistributedLEDPGraphAlgos

    2024. DistributedLEDPGraphAlgos. https://github.com/mundrapranay/ DistributedLEDPGraphAlgos

    2024

  2. [2]

    John M Abowd, Robert Ashmead, Ryan Cumings-Menon, Simson Garfinkel, Micah Heineck, Christine Heiss, Robert Johns, Daniel Kifer, Philip Leclerc, Ashwin Machanavajjhala, et al. 2022. The 2020 census disclosure avoidance system topdown algorithm. Harvard Data Science Review 2 (2022)

    2022

  3. [3]

    Jun Ai, Yayun Liu, Zhan Su, Fengyu Zhao, and Dunlu Peng. 2021. K-core decomposition in recommender systems improves accuracy of rating prediction. International Journal of Modern Physics C 32, 07 (2021), 2150087. https://doi. org/10.1142/S012918312150087X

    work page doi:10.1142/s012918312150087x 2021

  4. [4]

    Mohammad Al Hasan and Vachik S Dave. 2018. Triangle counting in large networks: a review. Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery 8, 2 (2018), e1226

    2018

  5. [5]

    Noga Alon, Raphael Yuster, and Uri Zwick. 1997. Finding and counting given length cycles. Algorithmica 17, 3 (1997), 209–223

    1997

  6. [6]

    Victor Balcer and Salil P. Vadhan. 2018. Differential Privacy on Finite Computers. In 9th Innovations in Theoretical Computer Science Conference (ITCS). 43:1–43:21

    2018

  7. [7]

    Bradley R Bebee, Daniel Choi, Ankit Gupta, Andi Gutmans, Ankesh Khandelwal, Yigit Kiran, Sainath Mallidi, Bruce McGaughy, Mike Personick, Karthik Rajan, et al. 2018. Amazon Neptune: Graph Data Management in the Cloud.. In ISWC (P&D/Industry/BlueSky). 15

    2018

  8. [8]

    Arijit Bishnu, Debarshi Chanda, and Gopinath Mishra. 2025. Arboricity and Random Edge Queries Matter for Triangle Counting using Sublinear Queries. arXiv:2502.15379 [cs.DS] https://arxiv.org/abs/2502.15379

    arXiv 2025

Show all 109 references
  1. [9]

    Arijit Bishnu, Debarshi Chanda, and Gopinath Mishra. 2025. Arboricity and Random Edge Queries Matter for Triangle Counting using Sublinear Queries. CoRR abs/2502.15379 (February 2025). https://doi.org/10.48550/arXiv.2502. 15379

    2025 doi arXiv

  2. [10]

    Francesco Bonchi, Aristides Gionis, and Francesco Gullo. 2014. Core decompo- sition of uncertain graphs. In Proceedings of the 20th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining . 1316–1325

    2014

  3. [11]

    Felipe T Brito, Victor AE Farias, Cheryl Flynn, Subhabrata Majumdar, Javam C Machado, and Divesh Srivastava. 2023. Global and local differentially private release of count-weighted graphs. Proceedings of the ACM on Management of Data 1, 2 (2023), 1–25

    2023

  4. [12]

    Mark Bun and Thomas Steinke. 2016. Concentrated Differential Privacy: Simpli- fications, Extensions, and Lower Bounds. In International Conference on Theory of Cryptography. 635–658

    2016

  5. [13]

    Hubert Chan, Elaine Shi, and Dawn Song

    T.-H. Hubert Chan, Elaine Shi, and Dawn Song. 2011. Private and Continual Release of Statistics. ACM Trans. Inf. Syst. Secur. 14, 3, Article 26 (Nov. 2011), 24 pages. https://doi.org/10.1145/2043621.2043626

    2011

  6. [14]

    Ho-Chun Herbert Chang and Emilio Ferrara. 2022. Comparative analysis of so- cial bots and humans during the COVID-19 pandemic.Journal of Computational Social Science 5, 2 (2022), 1409–1425

    2022

  7. [15]

    Martino Ciaperoni, Edoardo Galimberti, Francesco Bonchi, Ciro Cattuto, Francesco Gullo, and Alain Barrat. 2020. Relevance of temporal cores for epidemic spread in temporal networks. Scientific reports 10, 1 (2020), 12529

    2020

  8. [16]

    Camil Demetrescu, Andrew V Goldberg, David S Johnson, et al. 2008. Imple- mentation challenge for shortest paths. In Encyclopedia of Algorithms. Springer US, 395–398

    2008

  9. [17]

    Blelloch, and Julian Shun

    Laxman Dhulipala, Guy E. Blelloch, and Julian Shun. 2017. Julienne: A Frame- work for Parallel Graph Algorithms Using Work-efficient Bucketing. In ACM Symposium on Parallelism in Algorithms and Architectures (SPAA) . 293–304

    2017

  10. [18]

    Li, and Quanquan C

    Laxman Dhulipala, George Z. Li, and Quanquan C. Liu. 2024. Near-Optimal Differentially Private k-Core Decomposition. arXiv:2312.07706 [cs.DS] https: //arxiv.org/abs/2312.07706

    2024 arXiv

  11. [19]

    Liu, Sofya Raskhodnikova, Jessica Shi, Julian Shun, and Shangdi Yu

    Laxman Dhulipala, Quanquan C. Liu, Sofya Raskhodnikova, Jessica Shi, Julian Shun, and Shangdi Yu. 2022. Differential Privacy from Locally Adjustable Graph Algorithms: k-Core Decomposition, Low Out-Degree Ordering, and Densest Subgraphs. In 63rd IEEE Annual Symposium on Foundat...

    2022

  12. [20]

    Michael Dinitz, Satyen Kale, Silvio Lattanzi, and Sergei Vassilvitskii

  13. [21]

    Donovan and Brian W

    Alan A.A. Donovan and Brian W. Kernighan. 2015. The Go Programming Language (1st ed.). Addison-Wesley Professional

    2015

  14. [22]

    Cynthia Dwork and Jing Lei. 2009. Differential Privacy and Robust Statistics. In Proceedings of the Forty-First Annual ACM Symposium on Theory of Computing . 371–380

    2009

  15. [23]

    Cynthia Dwork, Frank McSherry, Kobbi Nissim, and Adam Smith. 2006. Cali- brating Noise to Sensitivity in Private Data Analysis. In Proceedings of the Third Conference on Theory of Cryptography . 265–284

    2006

  16. [24]

    Rothblum

    Cynthia Dwork, Moni Naor, Toniann Pitassi, and Guy N. Rothblum. 2010. Differ- ential Privacy under Continual Observation. In Proceedings of the Forty-Second ACM Symposium on Theory of Computing . 715–724

    2010

  17. [25]

    Rothblum, and Salil Vadhan

    Cynthia Dwork, Guy N. Rothblum, and Salil Vadhan. 2010. Boosting and Differ- ential Privacy. In Proceedings of the IEEE 51st Annual Symposium on Foundations of Computer Science. 51–60

    2010

  18. [26]

    Liu, Sofya Raskhodnikova, and Adam Smith

    Talya Eden, Quanquan C. Liu, Sofya Raskhodnikova, and Adam Smith. 2023. Triangle Counting with Local Edge Differential Privacy. In 50th International Colloquium on Automata, Languages, and Programming (ICALP 2023) (Leibniz International Proceedings in Informatics (LIPIcs), Vol...

    2023 doi

  19. [27]

    Liu, Sofya Raskhodnikova, and Adam D

    Talya Eden, Quanquan C. Liu, Sofya Raskhodnikova, and Adam D. Smith. 2023. Triangle Counting with Local Edge Differential Privacy. In 50th International Colloquium on Automata, Languages, and Programming, ICALP 2023, July 10-14, 2023, Paderborn, Germany (LIPIcs, Vol. 261) , Ko...

    2023 doi

  20. [28]

    Alexandre Evfimievski, Johannes Gehrke, and Ramakrishnan Srikant. 2003. Limiting privacy breaches in privacy preserving data mining. In Proceedings of the twenty-second ACM SIGMOD-SIGACT-SIGART symposium on Principles of database systems. 211–222

    2003

  21. [29]

    Alireza Farhadi, MohammadTaghi Hajiaghayi, and Elaine Shi. 2021. Differ- entially Private Densest Subgraph. CoRR abs/2106.00508 (2021), 11581–11597. arXiv:2106.00508 https://arxiv.org/abs/2106.00508

    2021 arXiv

  22. [30]

    Victor AE Farias, Felipe T Brito, Cheryl Flynn, Javam C Machado, Subhabrata Majumdar, and Divesh Srivastava. 2020. Local dampening: Differential privacy for non-numeric queries via local sensitivity. arXiv preprint arXiv:2012.04117 (2020)

    2020 arXiv

  23. [31]

    Nan Fu, Weiwei Ni, Sen Zhang, Lihe Hou, and Dongyue Zhang. 2023. GC- NLDP: A graph clustering algorithm with local differential privacy. Computers & Security 124 (2023), 102967

    2023

  24. [32]

    Kayo Fujimoto, Dimitrios Paraskevis, Jacky C Kuo, Camden J Hallmark, Jing Zhao, Andre Hochi, Lisa M Kuhns, Lu-Yu Hwang, Angelos Hatzakis, and John A Schneider. 2022. Integrated molecular and affiliation network analysis: Core- periphery social clustering is associated with HIV...

    2022

  25. [33]

    Christos Giatsidis, Fragkiskos D Malliaros, Nikolaos Tziortziotis, Charanpal Dhanjal, Emmanouil Kiagias, Dimitrios M Thilikos, and Michalis Vazirgiannis

  26. [34]

    Taolin Guo, Shunshun Peng, Yong Li, Mingliang Zhou, and Trieu-Kien Truong

  27. [35]

    Yang Guo, Fatemeh Esfahani, Xiaojian Shao, Venkatesh Srinivasan, Alex Thomo, Li Xing, and Xuekui Zhang. 2022. Integrative COVID-19 biological network inference with probabilistic core decomposition. Briefings in Bioinformatics 23, 1 (2022), bbab455

    2022

  28. [36]

    Jonathan Hehir, Aleksandra Slavković, and Xiaoyue Niu. 2022. Consistent spectral clustering of network block models under local differential privacy. The Journal of privacy and confidentiality 12, 2 (2022)

    2022

  29. [37]

    Monika Henzinger, A. R. Sricharan, and Leqi Zhu. 2024. Tighter Bounds for Local Differentially Private Core Decomposition and Densest Subgraph. CoRR abs/2402.18020 (2024). https://doi.org/10.48550/ARXIV.2402.18020 arXiv:2402.18020

    2024 doi arXiv

  30. [38]

    Seira Hidano and Takao Murakami. 2022. Degree-preserving randomized re- sponse for graph neural networks under local differential privacy.arXiv preprint arXiv:2202.10209 (2022)

    2022 arXiv

  31. [39]

    Quentin Hillebrand, Vorapong Suppakitpaisarn, and Tetsuo Shibuya. 2023. Com- munication Cost Reduction for Subgraph Counting under Local Differential Privacy via Hash Functions. arXiv preprint arXiv:2312.07055 (2023)

    2023 arXiv

  32. [40]

    Quentin Hillebrand, Vorapong Suppakitpaisarn, and Tetsuo Shibuya. 2023. Un- biased locally private estimator for polynomials of laplacian variables. In Pro- ceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. 741–751

    2023

  33. [41]

    Petter Holme and Nelly Litvak. 2017. Cost-efficient vaccination protocols for network epidemiology. PLoS computational biology 13, 9 (2017), e1005696

    2017

  34. [42]

    Jacob Imola, Alessandro Epasto, Mohammad Mahdian, Vincent Cohen-Addad, and Vahab Mirrokni. 2023. Differentially private hierarchical clustering with provable approximation guarantees. In International Conference on Machine Learning. PMLR, 14353–14375

    2023

  35. [43]

    Jacob Imola, Takao Murakami, and Kamalika Chaudhuri. 2021. Locally Differen- tially Private Analysis of Graph Statistics. In 30th USENIX Security Symposium . 983–1000

    2021

  36. [44]

    Jacob Imola, Takao Murakami, and Kamalika Chaudhuri. 2022. Communication- Efficient Triangle Counting under Local Differential Privacy. In 31st USENIX Security Symposium. 537–554

    2022

  37. [45]

    Linyu Jiang, Yukun Yan, Zhihong Tian, Zuobin Xiong, and Qilong Han. 2023. Personalized sampling graph collection with local differential privacy for link prediction. World Wide Web 26, 5 (2023), 2669–2689

    2023

  38. [46]

    Shiva Prasad Kasiviswanathan, Homin K Lee, Kobbi Nissim, Sofya Raskhod- nikova, and Adam Smith. 2011. What can we learn privately? SIAM J. Comput. 40, 3 (2011), 793–826

    2011

  39. [47]

    Schaefer

    Muah Kim, Onur Günlü, and Rafael F. Schaefer. 2021. Federated Learning with Local Differential Privacy: Trade-Offs Between Privacy, Utility, and Com- munication. In ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) . 2650–2654...

    2021

  40. [48]

    Gallos, Shlomo Havlin, Fredrik Liljeros, Lev Muchnik, H

    Maksim Kitsak, Lazaros K. Gallos, Shlomo Havlin, Fredrik Liljeros, Lev Muchnik, H. Eugene Stanley, and Hernán A. Makse. 2010. Identification of influential spreaders in complex networks. Nature Physics 6, 11 (Nov. 2010), 888–893. https://doi.org/10.1038/nphys1746

    2010 doi

  41. [49]

    Tejas Kulkarni. 2019. Answering Range Queries Under Local Differential Privacy. In Proceedings of the 2019 International Conference on Management of Data (Amsterdam, Netherlands)(SIGMOD ’19). Association for Computing Machinery, New York, NY, USA, 1832–1834. https://doi.org/10...

    2019

  42. [50]

    Haewoon Kwak, Changhyun Lee, Hosung Park, and Sue Moon. 2010. What is Twitter, a Social Network or a News Media?. In www. 591–600

    2010

  43. [51]

    Jure Leskovec and Andrej Krevl. 2014. SNAP Datasets: Stanford Large Network Dataset Collection. (2014)

    2014

  44. [52]

    Xiaoguang Li, Ninghui Li, Wenhai Sun, Neil Zhenqiang Gong, and Hui Li. 2023. Fine-grained poisoning attack to local differential privacy protocols for mean and variance estimation. In 32nd USENIX Security Symposium (USENIX Security 16 23). 1739–1756

    2023

  45. [53]

    Wanyu Lin, Baochun Li, and Cong Wang. 2022. Towards private learning on decentralized graphs with local differential privacy. IEEE Transactions on Information Forensics and Security 17 (2022), 2936–2946

    2022

  46. [54]

    Fang Liu, Dong Wang, and Tian Yan. 2023. Some examples of privacy-preserving sharing of COVID-19 pandemic data with statistical utility evaluation. BMC Medical Research Methodology 23, 1 (2023), 120

    2023

  47. [55]

    Liu, Jessica Shi, Shangdi Yu, Laxman Dhulipala, and Julian Shun

    Quanquan C. Liu, Jessica Shi, Shangdi Yu, Laxman Dhulipala, and Julian Shun

  48. [56]

    Shang Liu, Yang Cao, Takao Murakami, Jinfei Liu, and Masatoshi Yoshikawa

  49. [57]

    Shang Liu, Yang Cao, Takao Murakami, and Masatoshi Yoshikawa. 2022. A crypto-assisted approach for publishing graph statistics with node local dif- ferential privacy. In 2022 IEEE International Conference on Big Data (Big Data) . IEEE, 5765–5774

    2022

  50. [58]

    Yuhan Liu, Tianhao Wang, Yixuan Liu, Hong Chen, and Cuiping Li. 2024. Edge- Protected Triangle Count Estimation under Relationship Local Differential Privacy. IEEE Transactions on Knowledge and Data Engineering (2024)

    2024

  51. [59]

    Yuhan Liu, Suyun Zhao, Yixuan Liu, Dan Zhao, Hong Chen, and Cuiping Li

  52. [60]

    Zheng, Zhou Zhao, Hongxia Yang, Kevin Chen-Chuan Chang, Minghui Wu, and Jing Ying

    Zemin Liu, Vincent W. Zheng, Zhou Zhao, Hongxia Yang, Kevin Chen-Chuan Chang, Minghui Wu, and Jing Ying. 2018. Subgraph-Augmented Path Embed- ding for Semantic User Search on Heterogeneous Social Network. InProceedings of the 2018 World Wide Web Conference(Lyon, France) (WWW ’...

    2018

  53. [61]

    arXiv preprint arXiv:2312.12938 (2023)

    CARGO: Crypto-Assisted Differentially Private Triangle Counting with- out Trusted Servers. arXiv preprint arXiv:2312.12938 (2023)

    2023 arXiv

  54. [62]

    Naoki Masuda, Michiko Sakaki, Takahiro Ezaki, and Takamitsu Watanabe. 2018. Clustering Coefficients for Correlation Networks. Frontiers in Neuroinformatics 12 (2018). https://doi.org/10.3389/fninf.2018.00007

    2018

  55. [63]

    Gang Mei, Jingzhi Tu, Lei Xiao, and Francesco Piccialli. 2021. An efficient graph clustering algorithm by exploiting k-core decomposition and motifs. Computers & Electrical Engineering 96 (2021), 107564. https://doi.org/10.1016/j. compeleceng.2021.107564

    2021

  56. [64]

    Tamara T Mueller, Dmitrii Usynin, Johannes C Paetzold, Daniel Rueckert, and Georgios Kaissis. 2022. SoK: Differential privacy on graph-structured data. arXiv preprint arXiv:2203.09205 (2022)

    2022 arXiv

  57. [65]

    In 2022 IEEE 38th International Conference on Data Engineering (ICDE)

    Collecting triangle counts with edge relationship local differential privacy. In 2022 IEEE 38th International Conference on Data Engineering (ICDE) . IEEE, 2008–2020

    2022

  58. [66]

    Mohammad Naseri, Jamie Hayes, and Emiliano De Cristofaro. 2022. Local and Central Differential Privacy for Robustness and Privacy in Federated Learning. In 29th Annual Network and Distributed System Security Symposium, NDSS 2022, San Diego, California, USA, April 24-28, 2022 ....

    2022

  59. [67]

    Pathum Chamikara Mahawaga Arachchige, Dongxi Liu, Seyit Camtepe, Surya Nepal, Marthie Grobler, Peter Bertok, and Ibrahim Khalil. 2022. Local Differential Privacy for Federated Learning. InEuropean Symposium on Research in Computer Security. Springer, 195–216

    2022

  60. [68]

    Dung Nguyen, Mahantesh Halappanavar, Venkatesh Srinivasan, and Anil Vul- likanti. 2024. Faster approximate subgraph counts with privacy. Advances in Neural Information Processing Systems 36 (2024)

    2024

  61. [69]

    Kobbi Nissim, Sofya Raskhodnikova, and Adam Smith. 2007. Smooth Sensitivity and Sampling in Private Data Analysis. InProceedings of the Thirty-Ninth Annual ACM Symposium on Theory of Computing . 75–84

    2007

  62. [70]

    Gergely Palla, Imre Derényi, Illés Farkas, and Tamás Vicsek. 2005. Uncovering the overlapping community structure of complex networks in nature and society. Nature 435, 7043 (2005), 814–818

    2005

  63. [71]

    Takao Murakami and Yuichi Sei. 2023. Automatic Tuning of Privacy Budgets in Input-Discriminative Local Differential Privacy. IEEE Internet of Things Journal (2023)

    2023

  64. [72]

    Lei Qin, Yidan Wang, Qiang Sun, Xiaomei Zhang, Ben-Chang Shia, Chengcheng Liu, et al. 2020. Analysis of the covid-19 epidemic transmission network in mainland china: K-core decomposition study.JMIR public health and surveillance 6, 4 (2020), e24291

    2020

  65. [73]

    Neo4j. 2012. Neo4j - The World’s Leading Graph Database. http://neo4j.org/

    2012

  66. [74]

    Ryan Rossi and Nesreen Ahmed. 2015. The network data repository with inter- active graph analytics and visualization. In Proceedings of the AAAI conference on artificial intelligence, Vol. 29

    2015

  67. [75]

    Edo Roth, Karan Newatia, Yiping Ma, Ke Zhong, Sebastian Angel, and Andreas Haeberlen. 2021. Mycelium: Large-scale distributed graph queries with differ- ential privacy. In Proceedings of the ACM SIGOPS 28th Symposium on Operating Systems Principles. 327–343

    2021

  68. [76]

    Higor S Monteiro, Shaojun Luo, Saulo DS Reis, Carles Igual, Antonio S Lima Neto, Matias Travizan, Jose Soares De Andrade Jr, Hernan Makse, et al

  69. [77]

    Arnau Prat-Pérez, David Dominguez-Sal, Josep M Brunat, and Josep-Lluis Larriba-Pey. 2012. Shaping communities out of triangles. In Proceedings of the ACM international Conference on Information and Knowledge Management . 1677–1681

    2012

  70. [78]

    Mohamed Seif, Dung Nguyen, Anil Vullikanti, and Ravi Tandon. 2022. Differen- tially private community detection for stochastic block models. arXiv preprint arXiv:2202.00636 (2022)

    2022 arXiv

  71. [79]

    Zhan Qin, Ting Yu, Yin Yang, Issa Khalil, Xiaokui Xiao, and Kui Ren. 2017. Gen- erating Synthetic Decentralized Social Graphs with Local Differential Privacy. In Proceedings of the 2017 ACM SIGSAC Conference on Computer and Communi- cations Security (Dallas, Texas, USA) (CCS ’...

    2017

  72. [80]

    Matteo Serafino, Higor S Monteiro, Shaojun Luo, Saulo DS Reis, Carles Igual, Antonio S Lima Neto, Matías Travizano, José S Andrade Jr, and Hernán A Makse

  73. [81]

    Hubert Chan, Eleanor Gilbert Rieffel, Richard Chow, and Dawn Song

    Elaine Shi, T.-H. Hubert Chan, Eleanor Gilbert Rieffel, Richard Chow, and Dawn Song. 2011. Privacy-Preserving Aggregation of Time-Series Data. InProceedings of the Network and Distributed System Security Symposium

    2011

  74. [82]

    Tsourakakis

    Konstantinos Sotiropoulos and Charalampos E. Tsourakakis. 2021. Triangle- Aware Spectral Sparsifiers and Community Detection. In Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining (Virtual Event, Singapore) (KDD ’21). Association for Computing M...

    2021

  75. [83]

    Sriganesh Srihari and Hon Leong. 2013. A survey of computational methods for protein complex prediction from protein interaction networks. Journal of bioinformatics and computational biology 11 (04 2013), 1230002. https://doi.org/ 10.1142/S021972001230002X

    2013 doi

  76. [84]

    Stephen B Seidman. 1983. Network structure and minimum degree. Social networks 5, 3 (1983), 269–287

    1983

  77. [85]

    Marco Sánchez-Aguayo, Luis Urquiza-Aguiar, and José Estrada-Jiménez. 2021. Fraud Detection Using the Fraud Triangle Theory and Data Mining Tech- niques: A Literature Review. Computers 10, 10 (2021). https://doi.org/10. 3390/computers10100121

    2021

  78. [86]

    Monteiro, Shaojun Luo, and Hernán A

    Matteo Serafino, Higor S. Monteiro, Shaojun Luo, and Hernán A. Makse. 2020. Project COVID19 K-core tracker . https://github.com/makselab/COVID19

    2020

  79. [87]

    Google Differential Privacy Team. [n. d.]. GitHub - google/differential-privacy: Google’s differential privacy libraries. — github.com. https://github.com/google/ differential-privacy. [Accessed 12-04-2024]

    2024

  80. [88]

    PLOS Computational Biology 18, 4 (2022), e1009865

    Digital contact tracing and network theory to stop the spread of COVID- 19 using big-data on human mobility geolocalization. PLOS Computational Biology 18, 4 (2022), e1009865

    2022

  81. [89]

    Tom Tseng, Laxman Dhulipala, and Julian Shun. 2021. Parallel index-based structural graph clustering and its approximation. In Proceedings of the Interna- tional Conference on Management of Data . 1851–1864

    2021

  82. [90]

    Songlei Wang, Yifeng Zheng, Xiaohua Jia, Qian Wang, and Cong Wang. 2023. MAGO: Maliciously Secure Subgraph Counting on Decentralized Social Graphs. IEEE Transactions on Information Forensics and Security (2023)

    2023

  83. [91]

    Tianhao Wang, Bolin Ding, Jingren Zhou, Cheng Hong, Zhicong Huang, Ninghui Li, and Somesh Jha. 2019. Answering Multi-Dimensional Analyti- cal Queries under Local Differential Privacy. In Proceedings of the 2019 Inter- national Conference on Management of Data (Amsterdam, Nethe...

    2019

  84. [92]

    Haipei Sun, Xiaokui Xiao, Issa Khalil, Yin Yang, Zhan Qin, Hui Wang, and Ting Yu. 2019. Analyzing subgraph statistics from extended local views with decen- tralized differential privacy. In Proceedings of the 2019 ACM SIGSAC conference on computer and communications security . 703–717

    2019

  85. [93]

    Zhuo Wang, Zhixiong Li, Jinxing Tu, and Jianqiang Huang. 2025. DYTC:Dynamic Graph Triangle Counting on GPU. In Proceedings of the 2024 8th International Conference on Algorithms, Computing and Systems (ICACS ’24) . Association for Computing Machinery, New York, NY, USA, 83–87....

    2025

  86. [94]

    Differential Privacy Team. 2017. Learning with Privacy at Scale — machinelearn- ing.apple.com. https://machinelearning.apple.com/research/learning-with- privacy-at-scale. [Accessed 10-04-2024]

    2017

  87. [95]

    Zihang Xiang, Tianhao Wang, and Di Wang. 2023. Preserving Node-level Privacy in Graph Neural Networks. arXiv preprint arXiv:2311.06888 (2023)

    2023

  88. [96]

    Emre Gursoy

    Ekin Tire and M. Emre Gursoy. 2024. Answering Spatial Density Queries Under Local Differential Privacy. IEEE Internet of Things Journal 11, 10 (2024), 17419–17436. https://doi.org/10.1109/JIOT.2024.3357570

    2024

  89. [97]

    Qingqing Ye, Haibo Hu, Man Ho Au, Xiaofeng Meng, and Xiaokui Xiao. 2020. Towards locally differentially private generic graph metric estimation. In 2020 IEEE 36th International Conference on Data Engineering (ICDE). IEEE, 1922–1925

    2020

  90. [98]

    Wei Zeng, An Zeng, Hao Liu, Ming-Sheng Shang, and Tao Zhou. 2014. Uncov- ering the information core in recommender systems. Scientific reports 4 (August 2014), 6140. https://doi.org/10.1038/srep06140

    2014 doi

  91. [99]

    Da Zhong, Ruotong Yu, Kun Wu, Xiuling Wang, Jun Xu, and Wendy Hui Wang

  92. [100]

    Yanling Wang, Qian Wang, Lingchen Zhao, and Cong Wang. 2023. Differential privacy in deep learning: Privacy and beyond. Future Generation Computer Systems (2023)

    2023

  93. [102]

    Stanley L Warner. 1965. Randomized response: A survey technique for elimi- nating evasive answer bias. J. Amer. Statist. Assoc. 60, 309 (1965), 63–69

    1965

  94. [104]

    Qingqing Ye, Haibo Hu, Man Ho Au, Xiaofeng Meng, and Xiaokui Xiao. 2020. LF-GDPR: A framework for estimating graph metrics with local differential privacy. IEEE Transactions on Knowledge and Data Engineering 34, 10 (2020), 4905–4920. 17

    2020

  95. [108]

    Proceedings on Privacy Enhancing Technologies (2023)

    Disparate Vulnerability in Link Inference Attacks against Graph Neural Networks. Proceedings on Privacy Enhancing Technologies (2023)

    2023

  96. [109]

    Xiaochen Zhu, Vincent YF Tan, and Xiaokui Xiao. 2023. Blink: Link Local Differential Privacy in Graph Neural Networks via Bayesian Estimation. In Pro- ceedings of the 2023 ACM SIGSAC Conference on Computer and Communications Security. 2651–2664. 18

    2023

  97. [2016]

    arXiv preprint arXiv:1607.02096 (2016)

    A k-core decomposition framework for graph clustering. arXiv preprint arXiv:1607.02096 (2016)

    2016 arXiv

  98. [2021]

    Bulletin of the American Physical Society 66 (2021)

    Superspreading k-cores at the center of Covid-19 pandemic persistence. Bulletin of the American Physical Society 66 (2021)

    2021

  99. [2022]

    In 34th ACM Symposium on Parallelism in Algorithms and Architectures

    Parallel Batch-Dynamic Algorithms for𝑘-Core Decomposition and Re- lated Graph Problems. In 34th ACM Symposium on Parallelism in Algorithms and Architectures. 191–204

  100. [2023]

    Information Sciences 639 (2023), 119002

    Community-based social recommendation under local differential privacy protection. Information Sciences 639 (2023), 119002

    2023

  101. [2024]

    arXiv:2308.10316 [cs.DS]

    Almost Tight Bounds for Differentially Private Densest Subgraph. arXiv:2308.10316 [cs.DS]

    arXiv

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.