pith. sign in

arxiv: 2509.23205 · v1 · submitted 2025-09-27 · ⚛️ physics.soc-ph · cs.SI· physics.data-an

Network Inequality through Preferential Attachment, Triadic Closure, and Homophily

Jan Bachmann , Samuel Martin-Gutierrez , Lisette Esp\'in-Noboa , Nicola Cinardi , Fariba Karimi This is my paper

Pith reviewed 2026-05-18 13:10 UTC · model grok-4.3

classification ⚛️ physics.soc-ph cs.SIphysics.data-an
keywords preferential attachmenthomophilytriadic closurenetwork growthgender disparitiescollaboration networkssocial inequalitynetwork models
0
0 comments X

The pith

A model called PATCH shows that gender disparities in physics and computer science networks emerge from the combined action of preferential attachment, moderate homophily, and triadic closure.

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

The paper introduces PATCH, a model that grows networks by combining preferential attachment to popular nodes, homophily that favors links between similar nodes, and triadic closure through mutual contacts. Simulations demonstrate how preferential attachment and homophily raise both group segregation and degree inequalities, while triadic closure raises overall degree inequality but lowers segregation and the gap in degrees between groups. When the model is applied to fifty years of real collaboration and citation data, it reproduces the persistent gender differences seen in physics and computer science by using strong preferential attachment, moderate gender homophily, and adjustable triadic closure. A sympathetic reader would care because the work ties specific linking rules to measurable group outcomes and identifies possible points for intervention.

Core claim

PATCH is a network growth model that combines preferential attachment, homophily, and triadic closure. Simulations establish that homophily and preferential attachment increase segregation and degree inequalities, while triadic closure has opposing effects: it increases overall degree inequality yet reduces segregation and between-group degree disparities. PATCH reproduces the gender disparities observed in fifty years of physics and computer science collaboration networks when preferential attachment is strong, gender homophily is moderate, and triadic closure operates at varying strengths.

What carries the argument

PATCH, a network growth model that integrates preferential attachment, homophily, and triadic closure to generate and explain group disparities in synthetic and empirical networks.

If this is right

  • Homophily and preferential attachment together increase both segregation and degree inequality between groups.
  • Triadic closure increases population-wide degree inequality while decreasing segregation and between-group degree disparities.
  • The observed gender disparities in physics and computer science networks can be reproduced by moderate gender homophily, strong preferential attachment, and suitable triadic closure.
  • Adjusting the relative strengths of the three mechanisms offers a route to reduce group disparities in network outcomes.

Where Pith is reading between the lines

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

  • Interventions that increase triadic closure in academic networks could reduce gender segregation while leaving overall degree inequality largely unchanged.
  • The model suggests that early differences in attachment and closure rates compound over decades, implying that timing matters for efforts to equalize outcomes.
  • Similar patterns may appear when the same mechanisms are applied to other group attributes such as institutional prestige or geographic location.

Load-bearing premise

The real-world gender disparities in these collaboration networks arise primarily from the three modeled mechanisms and can be matched by tuning their relative strengths without large contributions from unmodeled factors.

What would settle it

Finding no combination of preferential attachment strength, homophily level, and triadic closure rate in PATCH that matches the measured degree distributions, segregation levels, and between-group degree gaps in the actual physics and computer science networks would falsify the account.

Figures

Figures reproduced from arXiv: 2509.23205 by Fariba Karimi, Jan Bachmann, Lisette Esp\'in-Noboa, Nicola Cinardi, Samuel Martin-Gutierrez.

Figure 1
Figure 1. Figure 1: PATCH networks. (a) Network growth: A new node i joins the network and is assigned to the majority group with probability 1− fmin, or the minority otherwise. For its first link, i selects a target node globally, from all existing nodes (gray outline). Subsequently, i creates m−1 additional links. For each new link, i connects to a target either globally with probability 1−τ, as in the first link, or throug… view at source ↗
Figure 2
Figure 2. Figure 2: Homophily drives segregation. We vary the triadic closure probability τ and homophily h (x-axes of the wider and shorter subplots, respectively) and the link formation combinations (a–d) to measure the network segregation by the EI-index (y-axis). Negative values indicate segregation, neutral values indicate mixing, and positive values indicate outgroup linking. (a) Due to the group size imbalance, the net… view at source ↗
Figure 3
Figure 3. Figure 3: Preferential attachment drives degree inequality. (a) Measuring global degree inequality by the Gini coefficient, we identify the strongest effect by the presence of preferential attachment (PAH,U) and (PAH,PAH). In heterophily (h < 0.5), only few nodes retain most of the available links, even in the absence of preferential attachment. While the effect of triadic closure depends on the other mechanisms, it… view at source ↗
Figure 4
Figure 4. Figure 4: Inequity and its relationship with inequality. (a) We measure inequity as the capability of one group to accumulate more links than the other by the Mann-Whitney test statistic. Values below 0.5 indicate that the majority group are advantaged in degree visibility, while values above 0.5 indicate the opposite. Homophily drives inequity while unbiased triadic closure and the presence of preferential attachme… view at source ↗
Figure 5
Figure 5. Figure 5: Gender inequalities in scientific networks over five decades. Each subplot shows an aggregate network statistic summarizing observed network inequalities and informing the PATCH model parameter inference. (a) While the fraction of women as the minority group fmin is increasing consistently over time for all datasets, it remains well below parity. (b) Segregation as measured by the EI-index increases simult… view at source ↗
Figure 6
Figure 6. Figure 6: Likelihood-free inference of PATCH variant and parameters. (a) Euclidean distance between observed and simulated network statistics for the empirical networks (color) and the PATCH model variants over decades. The best model variant is chosen based on the minimum distance (marked by a star; see Methods). (PAH,PAH) performs best in all cases, indicating that both the global and triadic closure target select… view at source ↗
read the original abstract

Inequalities in social networks arise from linking mechanisms, such as preferential attachment (connecting to popular nodes), homophily (connecting to similar others), and triadic closure (connecting through mutual contacts). While preferential attachment mainly drives degree inequality and homophily drives segregation, their three-way interaction remains understudied. This gap limits our understanding of how network inequalities emerge. Here, we introduce PATCH, a network growth model combining the three mechanisms to understand how they create disparities among two groups in synthetic networks. Extensive simulations confirm that homophily and preferential attachment increase segregation and degree inequalities, while triadic closure has countervailing effects: conditional on the other mechanisms, it amplifies population-wide degree inequality while reducing segregation and between-group degree disparities. We demonstrate PATCH's explanatory potential on fifty years of Physics and Computer Science collaboration and citation networks exhibiting persistent gender disparities. PATCH accounts for these gender disparities with the joint presence of preferential attachment, moderate gender homophily, and varying levels of triadic closure. By connecting mechanisms to observed inequalities, PATCH shows how their interplay sustains group disparities and provides a framework for designing interventions that promote more equitable 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

1 major / 2 minor

Summary. The manuscript introduces the PATCH network growth model integrating preferential attachment, homophily, and triadic closure to examine how these mechanisms generate group disparities in degree distributions and segregation. Simulations establish the directional and interactive effects of each process, with triadic closure shown to amplify overall inequality while mitigating between-group disparities under moderate homophily. The model is then applied to fifty years of Physics and Computer Science collaboration and citation networks, claiming that the joint presence of preferential attachment, moderate gender homophily, and varying triadic closure accounts for the observed persistent gender disparities.

Significance. If the central claims hold, this work supplies a mechanistic account of how standard network formation rules interact to sustain group-level inequalities, with the countervailing role of triadic closure representing a clear advance over pairwise mechanism studies. The simulations provide reproducible directional evidence for the three-way interplay, and the framework offers a basis for testing interventions aimed at equitable network growth. Credit is due for the explicit generative process and the attempt to connect micro-level rules to macro disparities in real collaboration data.

major comments (1)
  1. [Real-network application section] Real-network application section: the claim that PATCH 'accounts for' the observed gender disparities in degree and segregation rests on selecting parameter combinations (homophily strength, triadic closure rate, preferential attachment exponent) that align with empirical patterns. No details are supplied on the estimation procedure, optimization criterion, goodness-of-fit statistics, cross-validation, or sensitivity to parameter perturbations, rendering it impossible to distinguish explanatory power from post-hoc calibration.
minor comments (2)
  1. [Abstract and Simulations] The abstract and simulation results section would benefit from explicit statements of the explored parameter ranges and the number of independent runs used to establish the reported qualitative effects.
  2. [Model definition] Notation for the two groups (e.g., gender labels) and the precise functional forms of the attachment and closure probabilities should be consolidated in a single definitions subsection to improve readability.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the constructive feedback and positive assessment of the manuscript's significance. We address the major comment on the real-network application section below.

read point-by-point responses

  1. Referee: [Real-network application section] Real-network application section: the claim that PATCH 'accounts for' the observed gender disparities in degree and segregation rests on selecting parameter combinations (homophily strength, triadic closure rate, preferential attachment exponent) that align with empirical patterns. No details are supplied on the estimation procedure, optimization criterion, goodness-of-fit statistics, cross-validation, or sensitivity to parameter perturbations, rendering it impossible to distinguish explanatory power from post-hoc calibration.

    Authors: We agree that the current version of the real-network application section lacks sufficient detail on parameter selection and fit assessment. The parameters were chosen to be consistent with ranges reported in prior studies of academic collaboration networks (for homophily and triadic closure) and with maximum-likelihood estimates of the preferential attachment exponent from the empirical degree distributions, then adjusted within those ranges to produce qualitatively similar levels of gender segregation and degree inequality. In the revised manuscript we will expand the section to: (i) cite the specific literature sources and estimation methods for each parameter, (ii) report the exact values used for each network, (iii) add quantitative goodness-of-fit comparisons (Kolmogorov-Smirnov distances for degree distributions and direct comparisons of segregation and group-degree metrics), and (iv) include a sensitivity analysis showing how modest perturbations around the chosen values affect the match to the data. These additions will make the calibration process transparent and allow readers to evaluate the strength of the explanatory claim. revision: yes

Circularity Check

0 steps flagged

No significant circularity in derivation chain

full rationale

The PATCH model is defined by combining three explicit generative mechanisms (preferential attachment, homophily, triadic closure). Simulations then derive directional effects on segregation and degree inequality directly from the update rules; these outcomes are not equivalent to the inputs by construction and do not rely on self-citation chains or fitted parameters for their validity. The real-network application selects parameter values to align synthetic outputs with observed gender disparities in Physics and CS data, serving as a sufficiency demonstration rather than a load-bearing prediction or first-principles derivation that reduces to the empirical patterns themselves. No quoted step equates a claimed result to its own fitted inputs or prior author work.

Axiom & Free-Parameter Ledger

3 free parameters · 2 axioms · 0 invented entities

The model rests on several tunable parameters whose values are chosen to reproduce observed disparities and on standard assumptions of network growth models; no independent evidence for the parameter values is supplied in the abstract.

free parameters (3)
  • homophily strength
    Set to 'moderate' to match gender segregation patterns in simulations and real data
  • triadic closure rate
    Allowed to vary across periods to fit observed between-group degree disparities
  • preferential attachment exponent
    Implicitly present to drive degree inequality
axioms (2)
  • domain assumption Network growth proceeds by sequential addition of nodes and edges governed by the three mechanisms
    Foundational modeling choice for PATCH
  • domain assumption Group membership (gender) is a fixed binary attribute that influences link formation
    Required for homophily term in the model

pith-pipeline@v0.9.0 · 5757 in / 1508 out tokens · 65730 ms · 2026-05-18T13:10:06.614290+00:00 · methodology

discussion (0)

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

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Forward citations

Cited by 1 Pith paper

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

  1. Combining opinion and structural similarity in link recommendations to counter extreme polarization

    cs.SI 2026-04 unverdicted novelty 5.0

    Weak structural similarity combined with strong opinion similarity in link recommendations prevents network fragmentation and favors moderate opinions under strong homophily.

Reference graph

Works this paper leans on

57 extracted references · 57 canonical work pages · cited by 1 Pith paper · 1 internal anchor

  1. [1]

    Networks: An Introduction (Oxford University Press, 2010)

    Newman, M. Networks: An Introduction (Oxford University Press, 2010)

    work page 2010

  2. [2]

    P., Druschel, P

    Mislove, A., Marcon, M., Gummadi, K. P., Druschel, P. & Bhattacharjee, B. Measurement and analysis of online social networks. In Proceedings of the 7th ACM SIGCOMM conference on Internet measurement , IMC ’07, 29–42, DOI: 10.1145/1298306.1298311 (Association for Computing Machinery, New York, NY , USA, 2007)

    work page doi:10.1145/1298306.1298311 2007

  3. [3]

    Burt, R. S. Structural Holes: The Social Structure of Competition (Harvard University Press, Cambridge, MA, 1995)

    work page 1995

  4. [4]

    Social capital: A theory of social structure and action , vol

    Lin, N. Social capital: A theory of social structure and action , vol. 19 (Cambridge university press, 2002)

    work page 2002

  5. [5]

    J., Sinatra, R

    Huang, J., Gates, A. J., Sinatra, R. & Barab´asi, A.-L. Historical comparison of gender inequality in scientific careers across countries and disciplines. Proc. Natl. Acad. Sci. 117, 4609–4616, DOI: 10.1073/pnas.1914221117 (2020)

    work page doi:10.1073/pnas.1914221117 2020

  6. [6]

    & Karimi, F

    Kong, H., Martin-Gutierrez, S. & Karimi, F. Influence of the first-mover advantage on the gender disparities in physics citations. Commun. Phys. 5, 1–11, DOI: 10.1038/s42005-022-00997-x (2022)

    work page doi:10.1038/s42005-022-00997-x 2022

  7. [7]

    & Karimi, F

    Zappal`a, C., Gallo, L., Bachmann, J., Battiston, F. & Karimi, F. Gender disparities in the dissemination and acquisition of scientific knowledge, DOI: 10.48550/arXiv.2407.17441 (2024). 2407.17441

    work page doi:10.48550/arxiv.2407.17441 2024

  8. [8]

    & Horv´at, E.- ´A

    V´as´arhelyi, O., Zakhlebin, I., Milojevi´c, S. & Horv´at, E.- ´A. Gender inequities in the online dissemination of scholars’ work. Proc. Natl. Acad. Sci. 118, DOI: 10.1073/pnas.2102945118 (2021)

    work page doi:10.1073/pnas.2102945118 2021

  9. [9]

    Barab \' a si and R

    Barab´asi, A.-L. & Albert, R. Emergence of scaling in random networks. Science 286, 509–512, DOI: 10.1126/science.286.5439.509 (1999)

    work page doi:10.1126/science.286.5439.509 1999

  10. [10]

    Merton, R. K. The matthew effect in science. Science 159, 56–63, DOI: 10.1126/science.159.3810.56 (1968)

    work page doi:10.1126/science.159.3810.56 1968

  11. [11]

    Newman, M. E. J. Clustering and preferential attachment in growing networks. Phys. Rev. E 64, 025102, DOI: 10.1103/PhysRevE.64.025102 (2001)

    work page doi:10.1103/physreve.64.025102 2001

  12. [12]

    Cumulative Advantage of Brokerage in Academia

    Bachmann, J., Esp´ın-Noboa, L., I˜niguez, G. & Karimi, F. Cumulative advantage of brokerage in academia (2024). 2407.11909

    work page internal anchor Pith review Pith/arXiv arXiv 2024

  13. [13]

    Newman, M. E. J. The structure of scientific collaboration networks. Proc. Natl. Acad. Sci. 98, 404–409, DOI: 10.1073/pnas.98.2.404 (2001)

    work page doi:10.1073/pnas.98.2.404 2001

  14. [14]

    & Shirali, A

    Abebe, R., Immorlica, N., Kleinberg, J., Lucier, B. & Shirali, A. On the effect of triadic closure on network segregation. In Proceedings of the 23rd ACM Conference on Economics and Computation , EC ’22, 249–284, DOI: 10.1145/3490486.3538322 (Association for Computing Machinery, New York, NY , USA, 2022). 13/28

    work page doi:10.1145/3490486.3538322 2022

  15. [15]

    & Cook, J

    McPherson, M., Smith-Lovin, L. & Cook, J. M. Birds of a feather: Homophily in social networks. Annu. Rev. Sociol. 27, 415–444, DOI: 10.1146/annurev.soc.27.1.415 (2001)

    work page doi:10.1146/annurev.soc.27.1.415 2001

  16. [16]

    & Strohmaier, M

    Karimi, F., G´enois, M., Wagner, C., Singer, P. & Strohmaier, M. Homophily influences ranking of minorities in social networks. Sci. Reports 8, 11077, DOI: 10.1038/s41598-018-29405-7 (2018)

    work page doi:10.1038/s41598-018-29405-7 2018

  17. [17]

    & Thurner, S

    Klimek, P. & Thurner, S. Triadic closure dynamics drives scaling laws in social multiplex networks. New J. Phys. 15, 063008, DOI: 10.1088/1367-2630/15/6/063008 (2013)

    work page doi:10.1088/1367-2630/15/6/063008 2013

  18. [18]

    & Kivel ¨a, M

    Ure˜na-Carri´on, J., Karimi, F., ´I˜niguez, G. & Kivel ¨a, M. Assortative and preferential attachment lead to core-periphery networks. Phys. Rev. Res. 5, 043287, DOI: 10.1103/PhysRevResearch.5.043287 (2023)

    work page doi:10.1103/physrevresearch.5.043287 2023

  19. [19]

    Lee, E. et al. Homophily and minority-group size explain perception biases in social networks. Nat. Hum. Behav. 3, 1078–1087, DOI: 10.1038/s41562-019-0677-4 (2019)

    work page doi:10.1038/s41562-019-0677-4 2019

  20. [20]

    & Karimi, F

    Esp´ın-Noboa, L., Wagner, C., Strohmaier, M. & Karimi, F. Inequality and inequity in network-based ranking and recommendation algorithms. Sci. Reports 12, 2012, DOI: 10.1038/s41598-022-05434-1 (2022)

    work page doi:10.1038/s41598-022-05434-1 2012

  21. [21]

    & Kivelä, M

    Asikainen, A., I˜niguez, G., Ure˜na-Carri´on, J., Kaski, K. & Kivel¨a, M. Cumulative effects of triadic closure and homophily in social networks. Sci. Adv. 6, eaax7310, DOI: 10.1126/sciadv.aax7310 (2020)

    work page doi:10.1126/sciadv.aax7310 2020

  22. [22]

    & Eliassi-Rad, T

    Laber, M., Dies, S., Ehlert, J., Klein, B. & Eliassi-Rad, T. Effects of higher-order interactions and homophily on information access inequality, DOI: 10.48550/arXiv.2506.00156 (2025). 2506.00156

    work page doi:10.48550/arxiv.2506.00156 2025

  23. [23]

    Sajjadi, S. et al. Structural inequalities exacerbate infection disparities. Sci. Reports 15, 9082, DOI: 10.1038/ s41598-025-91008-w (2025)

    work page 2025

  24. [24]

    & Kim, B

    Holme, P. & Kim, B. J. Growing scale-free networks with tunable clustering. Phys. Rev. E 65, 026107, DOI: 10.1103/PhysRevE.65.026107 (2002)

    work page doi:10.1103/physreve.65.026107 2002

  25. [25]

    K., Iacovacci, J

    Bianconi, G., Darst, R. K., Iacovacci, J. & Fortunato, S. Triadic closure as a basic generating mechanism of communities in complex networks. Phys. Rev. E 90, 042806, DOI: 10.1103/PhysRevE.90.042806 (2014)

    work page doi:10.1103/physreve.90.042806 2014

  26. [26]

    & Schaub, M

    Neuh¨auser, L., Karimi, F., Bachmann, J., Strohmaier, M. & Schaub, M. T. Improving the visibility of minorities through network growth interventions. Commun. Phys. 6, 1–13, DOI: 10.1038/s42005-023-01218-9 (2023)

    work page doi:10.1038/s42005-023-01218-9 2023

  27. [27]

    & Watts, D

    Kossinets, G. & Watts, D. J. Origins of homophily in an evolving social network. Am. J. Sociol. 115, 405–450, DOI: 10.1086/599247 (2009)

    work page doi:10.1086/599247 2009

  28. [28]

    & Jackson, M

    Calv´o-Armengol, A. & Jackson, M. O. The effects of social networks on employment and inequality. Am. Econ. Rev. 94, 426–454, DOI: 10.1257/0002828041464542 (2004)

    work page doi:10.1257/0002828041464542 2004

  29. [29]

    & Karimi, F

    Sajjadi, S., Martin-Gutierrez, S. & Karimi, F. Unveiling homophily beyond the pool of opportunities, DOI: 10.48550/arXiv.2401.13642 (2024). 2401.13642

    work page doi:10.48550/arxiv.2401.13642 2024

  30. [30]

    & Stern, R

    Krackhardt, D. & Stern, R. N. Informal networks and organizational crises: An experimental simulation. Soc. Psychol. Q. 51, 123–140, DOI: 10.2307/2786835 (1988)

    work page doi:10.2307/2786835 1988

  31. [31]

    Feld, S. L. Why your friends have more friends than you do. Am. J. Sociol. 96, 1464–1477 (1991)

    work page 1991

  32. [32]

    McGraw, K. O. & Wong, S. P. A common language effect size statistic. Psychol. Bull. 111, 361–365, DOI: 10.1037/0033-2909.111.2.361 (1992)

    work page doi:10.1037/0033-2909.111.2.361 1992

  33. [33]

    Mann, H. B. & Whitney, D. R. On a test of whether one of two random variables is stochastically larger than the other. The Annals Math. Stat. 18, 50–60, DOI: 10.1214/aoms/1177730491 (1947)

    work page doi:10.1214/aoms/1177730491 1947

  34. [34]

    American physical society (aps) data sets for research (2023)

    American Physical Society. American physical society (aps) data sets for research (2023)

    work page 2023

  35. [35]

    dblp computer science bibliography – monthly snapshot xml release, DOI: 10.4230/dblp.xml (2016)

    dblp Team. dblp computer science bibliography – monthly snapshot xml release, DOI: 10.4230/dblp.xml (2016)

    work page doi:10.4230/dblp.xml 2016

  36. [36]

    Barnes, K. et al. Edge interventions can mitigate demographic and prestige disparities in the computer science coauthorship network, DOI: 10.48550/arXiv.2506.04435 (2025). 2506.04435. 14/28

    work page doi:10.48550/arxiv.2506.04435 2025

  37. [37]

    & Oliveira, M

    Karimi, F. & Oliveira, M. On the inadequacy of nominal assortativity for assessing homophily in networks. Sci. Reports 13, 21053, DOI: 10.1038/s41598-023-48113-5 (2023)

    work page doi:10.1038/s41598-023-48113-5 2023

  38. [38]

    & Wagner, C

    Jadidi, M., Karimi, F., Lietz, H. & Wagner, C. Gender disparities in science? dropout, productivity, col- laborations and success of male and female computer scientists. Adv. Complex Syst. 21, 1750011, DOI: 10.1142/S0219525917500114 (2017)

    work page doi:10.1142/s0219525917500114 2017

  39. [39]

    & Sugimoto, C

    Larivi`ere, V ., Ni, C., Gingras, Y ., Cronin, B. & Sugimoto, C. R. Bibliometrics: Global gender disparities in science. Nature 504, 211–213, DOI: 10.1038/504211a (2013)

    work page doi:10.1038/504211a 2013

  40. [40]

    Li, W., Zhang, S., Zheng, Z., Cranmer, S. J. & Clauset, A. Untangling the network effects of productivity and prominence among scientists. Nat. Commun. 13, 4907, DOI: 10.1038/s41467-022-32604-6 (2022)

    work page doi:10.1038/s41467-022-32604-6 2022

  41. [41]

    & Schweitzer, F

    Sarig¨ol, E., Pfitzner, R., Scholtes, I., Garas, A. & Schweitzer, F. Predicting scientific success based on coauthorship networks. EPJ Data Sci. 3, 9, DOI: 10.1140/epjds/s13688-014-0009-x (2014)

    work page doi:10.1140/epjds/s13688-014-0009-x 2014

  42. [42]

    The chaperone effect in scientific publishing

    Sekara, V .et al. The chaperone effect in scientific publishing. Proc. Natl. Acad. Sci. 115, 12603–12607, DOI: 10.1073/pnas.1800471115 (2018)

    work page doi:10.1073/pnas.1800471115 2018

  43. [43]

    & Eliassi-Rad, T

    Dies, S., Liu, D. & Eliassi-Rad, T. Forecasting faculty placement from patterns in co-authorship networks, DOI: 10.48550/arXiv.2507.14696 (2025). 2507.14696

    work page doi:10.48550/arxiv.2507.14696 2025

  44. [44]

    Dworkin, J. D. et al. The extent and drivers of gender imbalance in neuroscience reference lists. Nat. Neurosci. 23, 918–926, DOI: 10.1038/s41593-020-0658-y (2020)

    work page doi:10.1038/s41593-020-0658-y 2020

  45. [45]

    M., Macedo, M., Oliveira, M., Karimi, F

    Jaramillo, A. M., Macedo, M., Oliveira, M., Karimi, F. & Menezes, R. Systematic comparison of gender inequality in scientific rankings across disciplines, DOI: 10.48550/arXiv.2501.13061 (2025). 2501.13061

    work page doi:10.48550/arxiv.2501.13061 2025

  46. [46]

    Morgan, A. C. et al. The unequal impact of parenthood in academia. Sci. Adv. 7, eabd1996, DOI: 10.1126/ sciadv.abd1996 (2021)

    work page 2021

  47. [47]

    U., Cor, J

    Gutmann, M. U., Cor, J. & er. Bayesian optimization for likelihood-free inference of simulator-based statistical models. J. Mach. Learn. Res. 17, 1–47 (2016)

    work page 2016

  48. [48]

    Currarini, S., Jackson, M. O. & Pin, P. Identifying the roles of race-based choice and chance in high school friendship network formation. Proc. Natl. Acad. Sci. 107, 4857–4861, DOI: 10.1073/pnas.0911793107 (2010)

    work page doi:10.1073/pnas.0911793107 2010

  49. [49]

    V ., Clauset, A

    Buskirk, I. V ., Clauset, A. & Larremore, D. B. An open-source cultural consensus approach to name-based gender classification. Proc. Int. AAAI Conf. on Web Soc. Media 17, 866–877, DOI: 10.1609/icwsm.v17i1.22195 (2023)

    work page doi:10.1609/icwsm.v17i1.22195 2023

  50. [50]

    & Barabási, A.-L

    Sinatra, R., Wang, D., Deville, P., Song, C. & Barab´asi, A.-L. Quantifying the evolution of individual scientific impact. Science 354, aaf5239, DOI: 10.1126/science.aaf5239 (2016)

    work page doi:10.1126/science.aaf5239 2016

  51. [51]

    Rubin, D. B. Bayesianly justifiable and relevant frequency calculations for the applied statistician. The Annals Stat. 12, 1151–1172 (1984)

    work page 1984

  52. [52]

    U., Dutta, R., Kaski, S

    Lintusaari, J., Gutmann, M. U., Dutta, R., Kaski, S. & Corander, J. Fundamentals and recent developments in approximate bayesian computation. Syst. Biol. 66, e66–e82, DOI: 10.1093/sysbio/syw077 (2017)

    work page doi:10.1093/sysbio/syw077 2017

  53. [53]

    Marin, J.-M., Pudlo, P., Robert, C. P. & Ryder, R. J. Approximate bayesian computational methods. Stat. Comput. 22, 1167–1180, DOI: 10.1007/s11222-011-9288-2 (2012)

    work page doi:10.1007/s11222-011-9288-2 2012

  54. [54]

    Lintusaari, J. et al. Elfi: Engine for likelihood-free inference. J. Mach. Learn. Res. 19, 1–7 (2018)

    work page 2018

  55. [55]

    Adapting the abc distance function

    Prangle, D. Adapting the abc distance function. Bayesian Analysis 12, 289–309, DOI: 10.1214/16-BA1002 (2017)

    work page doi:10.1214/16-ba1002 2017

  56. [56]

    The dblp computer science bibliography: Evolution, research issues, perspectives

    Ley, M. The dblp computer science bibliography: Evolution, research issues, perspectives. In Laender, A. H. F. & Oliveira, A. L. (eds.) String Processing and Information Retrieval , 1–10, DOI: 10.1007/3-540-45735-6 1 (Springer, Berlin, Heidelberg, 2002). 15/28

    work page doi:10.1007/3-540-45735-6 2002

  57. [57]

    Internationale Kommunikation

    Kunegis, J. Konect: the koblenz network collection. In Proceedings of the 22nd International Conference on World Wide Web, WWW ’13 Companion, 1343–1350, DOI: 10.1145/2487788.2488173 (Association for Computing Machinery, New York, NY , USA, 2013). Acknowledgements The authors thank Eric Dignum for helpful discussions on the likelihood-free inference approa...

    work page doi:10.1145/2487788.2488173 2013