pith. sign in

arxiv: 2605.28772 · v1 · pith:7DV4DIZCnew · submitted 2026-05-27 · 💻 cs.SI

Sampling Random Graphs from the Colored Configuration Model

Leonardo Pellegrina This is my paper

Pith reviewed 2026-06-29 09:04 UTC · model grok-4.3

classification 💻 cs.SI
keywords colored configuration modelgraph null modelcolored degree matrixnetwork polarizationsocial network analysismetropolis-hastings samplingassortativityrandom graph generation
0
0 comments X

The pith

The Colored Configuration Model preserves each node's exact color-specific neighbor counts to enable controlled null-model tests of network features like polarization.

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

This paper introduces the Colored Configuration Model for vertex-colored multigraphs. The model keeps the Colored Degree Matrix fixed, recording precisely how many neighbors of each color every vertex possesses. Existing null models match only overall degrees and therefore leave color preferences free to vary. Fixing the matrix lets analysts check whether measured effects such as polarization are explained by those preferences alone or require additional network structure. Two Metropolis-Hastings samplers are supplied, one basic and one optimized to preserve the matrix with faster mixing, and experiments indicate that significance assessments can differ from those obtained with simpler null models.

Core claim

The Colored Configuration Model is a null model that exactly preserves the Colored Degree Matrix of an observed vertex-colored multigraph, thereby fixing the color assortativity of every node and permitting statistical tests that isolate the contribution of that matrix from other graph properties.

What carries the argument

The Colored Degree Matrix, which tabulates each vertex's neighbor counts by color, together with the Metropolis-Hastings procedures Sirius-B and Sirius that enforce its preservation during sampling.

If this is right

  • Significance tests for polarization measures can now be performed while holding node-level color assortativity constant.
  • Conclusions about whether a network feature requires explanations beyond color preferences may differ from those reached with degree-only null models.
  • The approach applies directly to online social networks in which colors mark opposing sides in debates.
  • Simpler null models that ignore the Colored Degree Matrix cannot isolate its effect on the same tasks.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same preservation idea could be extended to directed edges or to multiple simultaneous attributes beyond color.
  • Systematic application across many networks might reveal when color structure accounts for most of the signal in polarization or community measures.
  • The sampling machinery might be reused as a building block for null models that also fix other per-node matrices such as weighted interaction totals.

Load-bearing premise

That the color preferences recorded in the Colored Degree Matrix constitute the main confounding factor whose removal is required to judge whether other network traits drive an observed phenomenon.

What would settle it

A social network in which a polarization statistic is significant under degree-sequence null models but becomes insignificant (or vice versa) when sampled from the Colored Configuration Model would demonstrate that the matrix preservation changes the substantive conclusion.

Figures

Figures reproduced from arXiv: 2605.28772 by Leonardo Pellegrina.

Figure 2
Figure 2. Figure 2: Two colored multigraphs with the same degree [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 1
Figure 1. Figure 1: Two colored multigraphs with the same Colored [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 3
Figure 3. Figure 3: 𝑀 and 𝑀𝑣 on observed vs random graphs. We compare, for each graph 𝐺 ( [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Degree assortativity of Sirius and Sirius-B. out of space non-changing accepted rejected Sirius Sirius-B 0.0 0.2 0.4 0.6 0.8 1.0 proportion Phy-Cit Sirius Sirius-B 0.0 0.2 0.4 0.6 0.8 1.0 proportion Trivago Sirius Sirius-B 0.0 0.2 0.4 0.6 0.8 1.0 proportion Twitter [PITH_FULL_IMAGE:figures/full_fig_p008_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Iteration outcomes of Sirius and Sirius-B. To understand the reasons for this gap, we report in [PITH_FULL_IMAGE:figures/full_fig_p008_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Polarization scores on observed vs random graphs. [PITH_FULL_IMAGE:figures/full_fig_p009_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Values of 𝑀𝑣 for the 10 nodes with highest degree, on the observed graphs vs random samples [PITH_FULL_IMAGE:figures/full_fig_p013_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Degree assortativity of Sirius and Sirius-B up to 𝑚 ln𝑚 iterations [PITH_FULL_IMAGE:figures/full_fig_p014_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Degree assortativity of Sirius and Sirius-B up to 102𝑚 ln𝑚 iterations [PITH_FULL_IMAGE:figures/full_fig_p015_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: Iteration outcomes for Sirius and Sirius-B. 0 5000 10000 15000 20000 25000 30000 35000 40000 iteration 0.05 0.10 0.15 0.20 0.25 0.30 degree assortativity Convergence Cite Sirius Sirius-B Sirius CM Polaris Abortion Brexit Cite Com-DBLP Com-Youtube Comb Guns Obamacare Phy-Cit Trivago Twitter US-Elect Walmart 10 0 10 1 10 2 10 3 Running time (s) Running time (a) Abortion Brexit Cite Com-DBLP Com-Youtube Comb… view at source ↗
Figure 11
Figure 11. Figure 11: Running time comparison between Sirius and Sirius-B (a) and with CM and Polaris (b) [PITH_FULL_IMAGE:figures/full_fig_p016_11.png] view at source ↗
read the original abstract

A fundamental step in knowledge discovery is statistically assessing data mining results. In network analysis, such evaluation compares the outcome of a given procedure with the outcomes obtained from randomized versions of the observed network. Despite its importance, available graph null models only preserve simple characteristics of the observed graph, such as its degree sequence. In this paper we introduce the Colored Configuration Model (CCM), a new null model for vertex-colored multigraphs. Our main motivation is the study of online social networks, where the color of a user represents their side in a debate. The key novelty of CCM is preserving the Colored Degree Matrix (CDM), which encodes, for each vertex, the number of neighbors of any given color. Preserving the CDM allows fixing the color assortativity of all nodes, e.g., the propensity of each user to interact with other like-minded users. This allows testing whether a given phenomenon is explained by the observed CDM, or whether other characteristics of the network might play a key role. Available graph null models do not preserve the CDM, so they cannot assess its impact on real-world tasks, such as testing the significance of network polarization measures. To sample from the CCM, we develop Sirius-B, a simple baseline adapting the Metropolis-Hastings approach, and Sirius, a refined algorithm tailored to preserve the CDM, thus achieving provably faster mixing. In our experimental evaluation, we test Sirius on real-world networks, comparing it with related network null models. We observed that the evaluation of the statistical significance of polarization measures with Sirius may lead to different insights compared to available null models. Thus, Sirius is an effective tool for the statistically-sound analysis of social networks.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The paper introduces the Colored Configuration Model (CCM) as a null model for vertex-colored multigraphs that exactly preserves the Colored Degree Matrix (CDM) of an observed graph. This allows fixing per-vertex color assortativity. It proposes two MCMC samplers: Sirius-B (a Metropolis-Hastings baseline) and Sirius (a CDM-tailored algorithm with claimed provably faster mixing). Experiments on real-world networks compare Sirius to existing null models and report that CCM-based tests can yield different conclusions about the statistical significance of polarization measures.

Significance. If the preservation property and mixing guarantees hold, CCM fills a gap in network null models by enabling isolation of color-assortativity effects, which is directly relevant to analyzing polarization and echo chambers in social networks. The provision of reproducible sampling code and explicit comparison to baselines would strengthen its utility for downstream statistical testing.

major comments (2)
  1. [Theoretical analysis of mixing time (implied in abstract and algorithm sections)] The central claim that Sirius achieves provably faster mixing than the Metropolis-Hastings baseline requires explicit mixing-time bounds or coupling arguments; without these, the advantage over Sirius-B remains unverified for the CDM stationary distribution.
  2. [Experimental evaluation section] It is stated that available null models cannot assess CDM impact because they do not preserve it, but the manuscript should quantify the difference (e.g., via total variation distance or specific polarization statistic deviation) rather than only reporting qualitative changes in significance outcomes.
minor comments (2)
  1. Clarify the precise definition of the CDM (is it a per-vertex matrix or aggregated?) and whether the model permits self-loops or multiple edges between same-color pairs.
  2. Add a short pseudocode or step-by-step description of the Sirius proposal distribution to make the tailoring to CDM preservation transparent.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive review and the recommendation for minor revision. We address each major comment below.

read point-by-point responses

  1. Referee: The central claim that Sirius achieves provably faster mixing than the Metropolis-Hastings baseline requires explicit mixing-time bounds or coupling arguments; without these, the advantage over Sirius-B remains unverified for the CDM stationary distribution.

    Authors: We acknowledge that the manuscript asserts provably faster mixing for Sirius due to its CDM-specific design but does not supply explicit mixing-time bounds or a coupling argument in the current text. To address this, we will add a dedicated subsection in the theoretical analysis with a formal mixing-time comparison (via coupling or conductance bounds) between Sirius and Sirius-B under the CDM stationary distribution. revision: yes

  2. Referee: It is stated that available null models cannot assess CDM impact because they do not preserve it, but the manuscript should quantify the difference (e.g., via total variation distance or specific polarization statistic deviation) rather than only reporting qualitative changes in significance outcomes.

    Authors: We agree that quantitative comparisons would strengthen the experimental claims. In the revised experimental section we will report, for each real-world network, the absolute and relative deviations in the polarization statistics across null models, together with total-variation estimates between the empirical distributions generated by Sirius and the baseline models. revision: yes

Circularity Check

0 steps flagged

No significant circularity detected

full rationale

The paper defines the Colored Configuration Model explicitly by the property that it preserves the Colored Degree Matrix for each vertex. This is a modeling choice that directly encodes the desired null-model behavior rather than a derived prediction or result that reduces to its own inputs. The samplers (Sirius-B and Sirius) are constructed to sample from this defined distribution, with the faster mixing claim presented as a consequence of the tailoring to the CDM; no equations or steps in the abstract reduce a claimed result to a fitted parameter or self-citation by construction. The distinction from ordinary configuration models is stated explicitly and does not rely on any load-bearing self-reference or ansatz smuggling. The derivation chain is therefore self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 2 invented entities

Abstract-only review; no detailed derivation or parameter list visible. CCM appears to extend the standard configuration model by adding color constraints.

axioms (1)
  • standard math Standard configuration model assumptions for multigraphs apply to the colored extension
    CCM is described as a new null model building on configuration model ideas.
invented entities (2)
  • Colored Configuration Model (CCM) no independent evidence
    purpose: Null model that preserves Colored Degree Matrix
    New model introduced to address limitation of prior null models
  • Colored Degree Matrix (CDM) no independent evidence
    purpose: Matrix encoding per-vertex neighbor counts by color
    Core preserved quantity defining the model

pith-pipeline@v0.9.1-grok · 5823 in / 1101 out tokens · 36548 ms · 2026-06-29T09:04:40.649497+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. A Guide to Higher-Order Homophily

    physics.soc-ph 2026-06 unverdicted novelty 2.0

    A survey of existing measures and models for quantifying and generating higher-order homophily and heterophily in hypergraphs.

Reference graph

Works this paper leans on

61 extracted references · cited by 1 Pith paper

  1. [1]

    The Global Risks Report

    2024. The Global Risks Report. Technical Report.World Economic Forum(2024). https://www.weforum.org/publications/global-risks-report-2024

    2024

  2. [2]

    Florian Adriaens, Honglian Wang, and Aristides Gionis. 2023. Minimizing hitting time between disparate groups with shortcut edges. InProceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. 1–10

    2023

  3. [3]

    Ilya Amburg, Nate Veldt, and Austin Benson. 2020. Clustering in graphs and hypergraphs with categorical edge labels. InProceedings of the web conference

    2020

  4. [4]

    Albert-László Barabási and Réka Albert. 1999. Emergence of scaling in random networks.Science286, 5439 (1999), 509–512

    1999

  5. [5]

    Fabian Baumann, Philipp Lorenz-Spreen, Igor M Sokolov, and Michele Starnini

  6. [6]

    Physical review letters124, 4 (2020), 048301

    Modeling echo chambers and polarization dynamics in social networks. Physical review letters124, 4 (2020), 048301

    2020

  7. [7]

    Edward A Bender and E Rodney Canfield. 1978. The asymptotic number of labeled graphs with given degree sequences.Journal of Combinatorial Theory, Series A24, 3 (1978), 296–307

    1978

  8. [8]

    Béla Bollobás. 1980. A probabilistic proof of an asymptotic formula for the number of labelled regular graphs.European Journal of Combinatorics1, 4 (1980), 311–316

    1980

  9. [9]

    Duncan S Callaway, Mark EJ Newman, Steven H Strogatz, and Duncan J Watts

  10. [10]

    Network robustness and fragility: Percolation on random graphs.Physical review letters85, 25 (2000), 5468

    2000

  11. [11]

    Philip S Chodrow, Nate Veldt, and Austin R Benson. 2021. Generative hypergraph clustering: From blockmodels to modularity.Science Advances7, 28 (2021), eabh1303

    2021

  12. [12]

    Matteo Cinelli, Gianmarco De Francisci Morales, Alessandro Galeazzi, Walter Quattrociocchi, and Michele Starnini. 2021. The echo chamber effect on social media.Proceedings of the national academy of sciences118, 9 (2021), e2023301118

    2021

  13. [13]

    Federico Cinus, Aristides Gionis, and Francesco Bonchi. 2023. Rebalancing social feed to minimize polarization and disagreement. InProceedings of the 32nd ACM International Conference on Information and Knowledge Management. 369–378

    2023

  14. [14]

    Edward F Connor and Daniel Simberloff. 1979. The assembly of species commu- nities: chance or competition?Ecology60, 6 (1979), 1132–1140

    1979

  15. [15]

    Michael Conover, Jacob Ratkiewicz, Matthew Francisco, Bruno Gonçalves, Filippo Menczer, and Alessandro Flammini. 2011. Political polarization on twitter. In Proceedings of the international aaai conference on web and social media, Vol. 5. 89–96

    2011

  16. [16]

    Corinna Coupette, Stefan Neumann, and Aristides Gionis. 2023. Reducing ex- posure to harmful content via graph rewiring. InProceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. 323–334

    2023

  17. [17]

    Upasana Dutta, Bailey K Fosdick, and Aaron Clauset. 2025. Sampling random graphs with specified degree sequences.Journal of Computational and Graphical Statistics34, 3 (2025), 934–947

    2025

  18. [18]

    Paul Erdős and Alfréd Rényi. 1959. On Random Graphs I.Publicationes Mathe- maticae (Debrecen)6 (1959), 290–297

    1959

  19. [19]

    Francesco Fabbri, Yanhao Wang, Francesco Bonchi, Carlos Castillo, and Michael Mathioudakis. 2022. Rewiring what-to-watch-next recommendations to reduce radicalization pathways. InProceedings of the ACM Web Conference 2022. 2719– 2728

    2022

  20. [20]

    Ronald A. Fisher. 1935.The Design of Experiments. Oliver and Boyd, Edinburgh

    1935

  21. [21]

    Bailey K Fosdick, Daniel B Larremore, Joel Nishimura, and Johan Ugander. 2018. Configuring random graph models with fixed degree sequences.Siam Review60, 2 (2018), 315–355

    2018

  22. [22]

    Kiran Garimella, Gianmarco De Francisci Morales, Aristides Gionis, and Michael Mathioudakis. 2018. Political discourse on social media: Echo chambers, gate- keepers, and the price of bipartisanship. InProceedings of the 2018 world wide web conference. 913–922

    2018

  23. [23]

    Kiran Garimella, Aristides Gionis, Nikos Parotsidis, and Nikolaj Tatti. 2017. Bal- ancing information exposure in social networks.Advances in neural information processing systems30 (2017)

    2017

  24. [24]

    Kiran Garimella, Gianmarco De Francisci Morales, Aristides Gionis, and Michael Mathioudakis. 2017. The ebb and flow of controversial debates on social media. InProceedings of the International AAAI Conference on Web and Social Media, Vol. 11. 524–527

    2017

  25. [25]

    Kiran Garimella, Gianmarco De Francisci Morales, Aristides Gionis, and Michael Mathioudakis. 2018. Quantifying controversy on social media.ACM Transactions on Social Computing1, 1 (2018), 1–27

    2018

  26. [26]

    Sandra González-Bailón, David Lazer, Pablo Barberá, Meiqing Zhang, Hunt All- cott, Taylor Brown, Adriana Crespo-Tenorio, Deen Freelon, Matthew Gentzkow, Andrew M Guess, et al. 2023. Asymmetric ideological segregation in exposure to political news on Facebook.Science381, 6656 (2023), 392–398

    2023

  27. [27]

    Catherine Greenhill. 2014. The switch Markov chain for sampling irregular graphs. InProceedings of the twenty-sixth annual acm-siam symposium on discrete algorithms. SIAM, 1564–1572

    2014

  28. [28]

    Pedro Guerra, Wagner Meira Jr, Claire Cardie, and Robert Kleinberg. 2013. A measure of polarization on social media networks based on community bound- aries. InProceedings of the international AAAI conference on web and social media, Vol. 7. 215–224

    2013

  29. [29]

    Shahrzad Haddadan, Cristina Menghini, Matteo Riondato, and Eli Upfal. 2022. Reducing polarization and increasing diverse navigability in graphs by inserting edges and swapping edge weights.Data Mining and Knowledge Discovery36, 6 (2022), 2334–2378

    2022

  30. [30]

    Wilhelmiina Hämäläinen and Geoffrey I Webb. 2019. A tutorial on statistically sound pattern discovery.Data Mining and Knowledge Discovery33 (2019), 325– 377

    2019

  31. [31]

    Paul H Harvey, Robert K Colwell, Jonathan W Silvertown, and Robert M May

  32. [32]

    Null models in ecology.Annual Review of Ecology and Systematics14 (1983), 189–211

    1983

  33. [33]

    W Keith Hastings. 1970. Monte Carlo sampling methods using Markov chains and their applications. (1970)

    1970

  34. [34]

    Shalev Itzkovitz, Ron Milo, Nadav Kashtan, Guy Ziv, and Uri Alon. 2003. Sub- graphs in random networks.Physical review E68, 2 (2003), 026127

    2003

  35. [35]

    Lorien Jasny, Joseph Waggle, and Dana R Fisher. 2015. An empirical examination of echo chambers in US climate policy networks.Nature Climate Change5, 8 (2015), 782–786

    2015

  36. [36]

    Daniel B Larremore, Woodrow L Shew, and Juan G Restrepo. 2011. Predicting criticality and dynamic range in complex networks: effects of topology.Physical review letters106, 5 (2011), 058101

    2011

  37. [37]

    Antonis Matakos, Evimaria Terzi, and Panayiotis Tsaparas. 2017. Measuring and moderating opinion polarization in social networks.Data Mining and Knowledge Discovery31, 5 (2017), 1480–1505

    2017

  38. [38]

    Miller McPherson, Lynn Smith-Lovin, and James M Cook. 2001. Birds of a feather: Homophily in social networks.Annual review of sociology27, 1 (2001), 415–444

    2001

  39. [39]

    Nicholas Metropolis, Arianna W Rosenbluth, Marshall N Rosenbluth, Augusta H Teller, and Edward Teller. 1953. Equation of state calculations by fast computing machines.The journal of chemical physics21, 6 (1953), 1087–1092

    1953

  40. [40]

    Ron Milo, Shai Shen-Orr, Shalev Itzkovitz, Nadav Kashtan, Dmitri Chklovskii, and Uri Alon. 2002. Network motifs: simple building blocks of complex networks. Science298, 5594 (2002), 824–827

    2002

  41. [41]

    Antonietta Mira. 2001. Ordering and improving the performance of Monte Carlo Markov chains.Statist. Sci.(2001), 340–350

    2001

  42. [42]

    Gummadi, Peter Druschel, and Bobby Bhattacharjee

    Alan Mislove, Massimiliano Marcon, Krishna P. Gummadi, Peter Druschel, and Bobby Bhattacharjee. 2007. Measurement and Analysis of Online Social Networks. InProceedings of the 5th ACM/Usenix Internet Measurement Conference (IMC’07). San Diego, CA

    2007

  43. [43]

    2017.Probability and computing: Random- ization and probabilistic techniques in algorithms and data analysis

    Michael Mitzenmacher and Eli Upfal. 2017.Probability and computing: Random- ization and probabilistic techniques in algorithms and data analysis. Cambridge university press

    2017

  44. [44]

    Jacob L Moreno and Helen H Jennings. 1938. Statistics of social configurations. Sociometry(1938), 342–374

    1938

  45. [45]

    Mark EJ Newman. 2003. Mixing patterns in networks.Physical review E67, 2 (2003), 026126

    2003

  46. [46]

    Mark EJ Newman, Steven H Strogatz, and Duncan J Watts. 2001. Random graphs with arbitrary degree distributions and their applications.Physical review E64, 2 (2001), 026118

    2001

  47. [47]

    Leonardo Pellegrina, Matteo Riondato, and Fabio Vandin. 2019. Hypothesis Testing and Statistically-sound Pattern Mining. InProceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining (KDD ’19). ACM, New York, NY, USA, 3215–3216

    2019

  48. [48]

    Peter H Peskun. 1973. Optimum monte-carlo sampling using markov chains. Biometrika60, 3 (1973), 607–612

    1973

  49. [49]

    Giulia Preti, Gianmarco De Francisci Morales, and Matteo Riondato. 2023. Mani- acs: Approximate mining of frequent subgraph patterns through sampling.ACM Transactions on Intelligent Systems and Technology14, 3 (2023), 1–29

    2023

  50. [50]

    Giulia Preti, Matteo Riondato, and Aristides Gionis. 2026. DSP: A Statistically- Principled Structural Polarization Measure. InProceedings of the Nineteenth ACM International Conference on Web Search and Data Mining

    2026

  51. [51]

    Giulia Preti, Matteo Riondato, Aristides Gionis, and Gianmarco De Fran- cisci Morales. 2025. Polaris: Sampling from the Multigraph Configuration Model with Prescribed Color Assortativity. InProceedings of the Eighteenth ACM Inter- national Conference on Web Search and Data Mining. 30–39

    2025

  52. [52]

    Matteo Riondato. 2023. Statistically-sound knowledge discovery from data. In Proceedings of the 2023 SIAM International Conference on Data Mining (SDM). SIAM, 949–952

    2023

  53. [53]

    Vivekananda Roy. 2020. Convergence diagnostics for markov chain monte carlo. Annual Review of Statistics and Its Application7, 1 (2020), 387–412

    2020

  54. [54]

    Ali Salloum, Ted Hsuan Yun Chen, and Mikko Kivelä. 2022. Separating po- larization from noise: comparison and normalization of structural polarization measures.Proceedings of the ACM on human-computer interaction6, CSCW1 (2022), 1–33

    2022

  55. [55]

    Tom AB Snijders et al. 2002. Markov chain Monte Carlo estimation of exponential random graph models.Journal of Social Structure3, 2 (2002), 1–40

    2002

  56. [56]

    2017.Maximum-entropy networks: Pattern detection, network reconstruction and graph combinatorics

    Tiziano Squartini and Diego Garlaschelli. 2017.Maximum-entropy networks: Pattern detection, network reconstruction and graph combinatorics. Springer. Sampling Random Graphs from the Colored Configuration Model KDD ’26, August 09–13, 2026, Jeju Island, Republic of Korea

    2017

  57. [57]

    Arun Sundararajan. 2008. Local Network Effects and Complex Network Structure. The BE Journal of Theoretical Economics7, 1 (2008), 1–37

    2008

  58. [58]

    2016.Introduction to data mining

    Pang-Ning Tan, Michael Steinbach, and Vipin Kumar. 2016.Introduction to data mining. Pearson Education India

    2016

  59. [59]

    Jaewon Yang and Jure Leskovec. 2012. Defining and evaluating network commu- nities based on ground-truth. InProceedings of the ACM SIGKDD workshop on mining data semantics. 1–8

    2012

  60. [60]

    Liwang Zhu, Qi Bao, and Zhongzhi Zhang. 2021. Minimizing polarization and disagreement in social networks via link recommendation.Advances in Neural Information Processing Systems34 (2021), 2072–2084

    2021

  61. [61]

    {𝑐(𝑢), 𝑐(𝑣)}={𝑐(𝑥), 𝑐(𝑦)}={ℓ, 𝑟}

    Liwang Zhu and Zhongzhi Zhang. 2022. A nearly-linear time algorithm for minimizing risk of conflict in social networks. InProceedings of the 28th ACM SIGKDD conference on knowledge discovery and data mining. 2648–2656. A Appendix This appendix presents the proofs of our theoretical derivations, and experimental results that could not fit in the main manus...

    2022