- Physical Review Journals
All JournalsPhysics Magazine

Physical Review Research

Reuse & Permissions

It is not necessary to obtain permission to reuse this article or its components as it is available under the terms of the Creative Commons Attribution 4.0 International license. This license permits unrestricted use, distribution, and reproduction in any medium, provided attribution to the author(s) and the published article's title, journal citation, and DOI are maintained. Please note that some figures may have been included with permission from other third parties. It is your responsibility to obtain the proper permission from the rights holder directly for these figures.

Export citation

Export citation

Choose format for download:

Download Citation
  • Open Access

Missing links prediction: Comparing machine learning with physics-rooted approaches

Francesca Santucci1,*, Giulio Cimini2,3, and Tiziano Squartini1,4,5

  • *Contact author: santuccifrancesca26@gmail.com
PDF

Phys. Rev. Research 8, 043020 – Published 7 October, 2026

DOI: https://doi.org/10.1103/76kb-hyp4

Abstract

An active research line within the broader field of network science is the one concerning link prediction. Close in scope to network reconstruction, link prediction targets specific connections with the aim of uncovering the missing ones, as well as predicting those most likely to emerge in the future, from the available information. In this paper, we consider two families of methods, i.e., those rooted in statistical physics and those based upon machine learning: the members of the first family identify missing links as the most probable nonobserved ones, the probability coefficients being determined by solving maximum-entropy benchmarks over the accessible network structure; the members of the second family, instead, associate the presence of single edges with explanatory node-specific variables. Running likelihood-based models such as the configuration model, or one of its many fitness-based variants, in parallel with the gradient boosting decision tree algorithm reveals that the accuracy of the former is comparable to the accuracy of the latter. Such a result confirms that white-box algorithms are viable competitors to the currently available black-box ones, being more interpretable and computationally faster.

View figure in article

Physics Subject Headings (PhySH)

Article Text

References (52)

  1. T. Squartini, G. Caldarelli, G. Cimini, A. Gabrielli, and D. Garlaschelli, Reconstruction methods for networks: The case of economic and financial systems, Phys. Rep. 757, 1 (2018).
  2. L. Lü and T. Zhou, Link prediction in complex networks: A survey, Physica A 390, 1150 (2011).
  3. D. Liben-Nowell and J. Kleinberg, The link-prediction problem for social networks, J. Am. Soc. Inf. Sci. Technol. 58, 1019 (2007).
  4. G. Berlusconi, F. Calderoni, N. Parolini, M. Verani, and C. Piccardi, Link prediction in criminal networks: A tool for criminal intelligence analysis, PLoS One 11, e0154244 (2016).
  5. M. Jalili, Y. Orouskhani, M. Asgari, N. Alipourfard, and M. Perc, Link prediction in multiplex online social networks, R. Soc. Open Sci. 4, 160863 (2017).
  6. F. Parisi, G. Caldarelli, and T. Squartini, Entropy-based approach to missing-links prediction, Appl. Network Sci. 3, 17 (2018).
  7. L. Mungo, F. Lafond, P. Astudillo-Estévez, and J. D. Farmer, Reconstructing production networks using machine learning, J. Econ. Dyn. Control 148, 104607 (2023).
  8. M. Bardoscia, P. Barucca, S. Battiston, F. Caccioli, G. Cimini, D. Garlaschelli, F. Saracco, T. Squartini, and G. Caldarelli, The physics of financial networks, Nat. Rev. Phys. 3, 490 (2021).
  9. L. N. Ialongo, C. de Valk, E. Marchese, F. Jansen, H. Zmarrou, T. Squartini, and D. Garlaschelli, Reconstructing firm-level interactions in the Dutch input–output network from production constraints, Sci. Rep. 12, 11847 (2022).
  10. A. L. Barabási and R. Albert, Emergence of scaling in random networks, Science 286, 509 (1999).
  11. T. Zhou, L. Lü, and Y.-C. Zhang, Predicting missing links via local information, Eur. Phys. J. B 71, 623 (2009).
  12. L. Katz, A new status index derived from sociometric analysis, Psychometrika 18, 39 (1953).
  13. S. Brin and L. Page, The anatomy of a large-scale hypertextual Web search engine, Comput. Networks ISDN Syst. 30, 107 (1998).
  14. P. Y. Chebotarev and E. Shamis, The matrix-forest theorem and measuring relations in small social groups, Autom. Remote Control 58, 1505 (1997).
  15. G. Jeh and J. Widom, SimRank: A measure of structural-context similarity, in Proceedings of the Eighth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD '02) (Association for Computing Machinery, New York, NY, 2002), pp. 538–543.
  16. W. Liu and L. Lü, Link prediction based on local random walk, Europhys. Lett. 89, 58007 (2010).
  17. A. Clauset, C. Moore, and M. E. J. Newman, Hierarchical structure and the prediction of missing links in networks, Nature (London) 453, 98 (2008).
  18. R. Guimerà and M. Sales-Pardo, Missing and spurious interactions and the reconstruction of complex networks, Proc. Natl. Acad. Sci. USA 106, 22073 (2009).
  19. F. Adriaens, A. Mara, J. Lijffijt, and T. De Bie, Block-approximated exponential random graphs, in 2020 IEEE 7th International Conference on Data Science and Advanced Analytics (DSAA) (IEEE, Piscataway, NJ, 2020), pp. 70–80.
  20. N. Friedman, L. Getoor, D. Koller, and A. Pfeffer, Learning probabilistic relational models, in Proceedings of the 16th International Joint Conference on Artificial Intelligence (IJCAI'99) (Morgan Kaufmann, San Francisco, CA, 1999), Vol. 2, pp. 1300–1307.
  21. K. Yu, W. Chu, S. Yu, V. Tresp, and Z. Xu, Stochastic relational models for discriminative link prediction, in Advances in Neural Information Processing Systems 19, edited by B. Schölkopf, J. Platt, and T. Hofmann (MIT Press, Cambridge, MA, 2007), pp. 1553–1560.
  22. The terms “endogenous” and “exogenous” are borrowed from economics to distinguish between features that are structural in nature, like the degrees (“endogenous”, i.e., “inherent to the structure”) and features that, instead, are not, like the gross domestic products and the geographic distances (“exogenous”, i.e., “not inherent to the structure”).
  23. A. Polanco and M. E. J. Newman, Drug-disease networks and drug repurposing, PLoS Comput. Biol. 21, e1013595 (2025).
  24. V. Krishnamurthy, M. Faloutsos, M. Chrobak, L. Lao, J. H. Cui, and A. G. Percus, Reducing large internet topologies for faster simulations, in NETWORKING 2005. Networking Technologies, Services, and Protocols; Performance of Computer and Communication Networks; Mobile and Wireless Communications Systems, edited by D. Hutchison, T. Kanade, J. Kittler, J. M. Kleinberg, F. Mattern, J. C. Mitchell, M. Naor, O. Nierstrasz, C. Pandu Rangan, B. Steffen, M. Sudan, D. Terzopoulos, D. Tygar, M. Y. Vardi, G. Weikum, R. Boutaba, K. Almeroth, R. Puigjaner, S. Shen, and J. P. Black, Lecture Notes in Computer Science (Springer, Berlin, 2005), pp. 328–341.
  25. W. Isard, Location theory and trade theory: Short-run analysis, Q. J. Econ. 68, 305 (1954).
  26. J. Tinbergen, Shaping the World Economy: Suggestions for an International Economic Policy (Twentieth Century Fund, New York, 1962).
  27. F. Pedregosa, G. Varoquaux, A. Gramfort, V. Michel, B. Thirion, O. Grisel, M. Blondel, P. Prettenhofer, R. Weiss, V. Dubourg, J. VanderPlas, A. Passos, D. Cournapeau, M. Brucher, M. Perrot, and E. Duchesnay, Scikit-learn: Machine learning in Python, J. Mach. Learn. Res. 12, 2825 (2011).
  28. J. H. Friedman, Greedy function approximation: A gradient boosting machine, Ann. Stat. 29, 1189 (2001).
  29. G. Ke, Q. Meng, T. Finley, T. Wang, W. Chen, W. Ma, Q. Ye, and T.-Y. Liu, LightGBM: A highly efficient gradient boosting decision tree, in Advances in Neural Information Processing Systems 30, edited by I. Guyon, U. von Luxburg, S. Bengio, H. Wallach, R. Fergus, S. Vishwanathan, and R. Garnett (Curran Associates, Red Hook, NY, 2017), pp. 3146–3154.
  30. J. Park and M. E. J. Newman, Statistical mechanics of networks, Phys. Rev. E 70, 066117 (2004).
  31. G. Bianconi, The entropy of randomized network ensembles, Europhys. Lett. 81, 28005 (2008).
  32. T. Squartini and D. Garlaschelli, Analytical maximum-likelihood method to detect patterns in real networks, New J. Phys. 13, 083001 (2011).
  33. A. Fronczak, Statistical mechanics of the international trade network, Phys. Rev. E 85, 056113 (2012).
  34. T. Squartini, R. Mastrandrea, and D. Garlaschelli, Unbiased sampling of network ensembles, New J. Phys. 17, 023052 (2015).
  35. F. Saracco, R. Di Clemente, A. Gabrielli, and T. Squartini, Randomizing bipartite networks: The case of the World Trade Web, Sci. Rep. 5, 10595 (2015).
  36. G. Cimini, T. Squartini, F. Saracco, D. Garlaschelli, A. Gabrielli, and G. Caldarelli, The statistical physics of real-world networks, Nat. Rev. Phys. 1, 58 (2019).
  37. S. N. Dorogovtsev, J. F. F. Mendes, and A. N. Samukhin, Principles of statistical mechanics of uncorrelated random networks, Nucl. Phys. B 666, 396 (2003.
  38. J. Park and M. E. J. Newman, Origin of degree correlations in the Internet and other networks, Phys. Rev. E 68, 026112 (2003).
  39. D. Garlaschelli and M. I. Loffredo, Maximum likelihood: Extracting unbiased information from complex networks, Phys. Rev. E 78, 015101(R) (2008).
  40. N. Vallarano, M. Bruno, E. Marchese, G. Trapani, F. Saracco, G. Cimini, M. Zanon, and T. Squartini, Fast and scalable likelihood maximization for exponential random graph models with local constraints, Sci. Rep. 11, 15227 (2021).
  41. M. Di Vece, Deterministic, quenched, and annealed parameter estimation for heterogeneous network models, Phys. Rev. E 108, 054301 (2023).
  42. M. Marzi, F. Giuffrida, D. Garlaschelli, and T. Squartini, Reproducing the first and second moment of empirical degree distributions Phys. Rev. Res. 8, 013141 (2026).
  43. In a sense, the link prediction problem is interpreted as an instance of the network reconstruction one [1, 49], the aim of which is precisely that of inferring a graph structure from a set of observed quantities—although network reconstruction usually deals with less information to predict more aggregate properties [see also Fig.  1].
  44. F. Picciolo, T. Squartini, F. Ruzzenenti, R. Basosi, and D. Garlaschelli, The role of distances in the World Trade Web, in 2012 Eighth International Conference on Signal Image Technology and Internet Based Systems (SITIS) (IEEE Computer Society, Los Alamitos, CA, 2012), pp. 784–792.
  45. G. Bianconi, Information theory of spatial network ensembles, in Handbook on Entropy, Complexity and Spatial Dynamics (Edward Elgar Publishing, Cheltenham, UK, 2021), pp. 61–96.
  46. G. Caldarelli, A. Capocci, P. De Los Rios, and M. A. Muñoz, Scale-free networks from varying vertex intrinsic fitness, Phys. Rev. Lett. 89, 258702 (2002).
  47. D. Garlaschelli and M. I. Loffredo, Fitness-dependent topological properties of the World Trade Web, Phys. Rev. Lett. 93, 188701 (2004).
  48. K. S. Gleditsch, Expanded trade and GDP data, J. Conflict Resolut. 46, 712 (2002).
  49. G. Cimini, R. Mastrandrea, and T. Squartini, Reconstructing Networks (Cambridge University Press, Cambridge, 2021).
  50. G. Cimini, T. Squartini, D. Garlaschelli, and A. Gabrielli, Systemic risk analysis on reconstructed economic and financial networks, Sci. Rep. 5, 15758 (2015).
  51. J. Park and M. E. J. Newman, Solution of the two-star model of a network, Phys. Rev. E 70, 066146 (2004).
  52. M. Bolfe, F. L. Metz, E. Guzmán-González, and I. P. Castillo, Analytic solution of the two-star model with correlated degrees, Phys. Rev. E 104, 014147 (2021).

Outline

Information

Phys. Rev. Research 8, 043020– Published 7 October, 2026
  • Received 5 February 2026
  • Accepted 4 August 2026
Crossmark - Check for updates button
Creative Commons Logo

Published by the American Physical Society under the terms of the Creative Commons Attribution 4.0 International license. Further distribution of this work must maintain attribution to the author(s) and the published article's title, journal citation, and DOI.

Published by the American Physical Society

Sign In to Your Journals Account

Filter

Filter

Article Lookup

Enter a citation