Model-free inference of diffusion networks using RKHS embeddings

Abstract

We revisit in this paper the problem of inferring a diffusion network from information cascades. In our study, we make no assumptions on the underlying diffusion model, in this way obtaining a generic method with broader practical applicability. Our approach exploits the pairwise adoption-time intervals from cascades. Starting from the observation that different kinds of information spread differently, these time intervals are interpreted as samples drawn from unknown (conditional) distributions. In order to statistically distinguish them, we propose a novel method using Reproducing Kernel Hilbert Space embeddings. Experiments on both synthetic and real-world data from Twitter and Flixster show that our method significantly outperforms the state-of-the-art methods. We argue that our algorithm can be implemented by parallel batch processing, in this way meeting the needs in terms of efficiency and scalability of real-world applications.

This is a preview of subscription content, log in to check access.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6

Notes

  1. 1.

    E.g., we set 20 as a minimal support threshold in our experiments, since inference on candidate edges with very few adoption time intervals would have little statistical significance.

  2. 2.

    Superscript (uv) on \(y_{c}\) is omitted as it remains the same for all time intervals in \(D^{(i)}\).

  3. 3.

    To support this intuition, we also show in Fig. 1 the distribution of adoption time intervals conditioned by movie popularity in our Flixster experimental dataset, for connected (blue) or unconnected (red) user pairs. For similar empirical evidence we also refer the reader to Du et al. (2013).

  4. 4.

    http://snap.stanford.edu/.

  5. 5.

    We are grateful to the authors for the binary package of NPDC and one synthetic dataset.

  6. 6.

    https://github.com/amber0309/KEBC.

  7. 7.

    The proportion of pairs with edge and without edge in these batches should be very similar to the one of the entire dataset.

  8. 8.

    http://imdbpy.sourceforge.net.

References

  1. Arthur D, Vassilvitskii S (2007) K-means++: The advantages of careful seeding. In: Proceedings of the eighteenth annual ACM-SIAM symposium on discrete algorithms, SODA ’07, pp 1027–1035, Philadelphia. Society for Industrial and Applied Mathematics

  2. Chen W, Wang Y, Yuan Y, Wang Q (2016) Combinatorial multi-armed bandit and its extension to probabilistically triggered arms. J Mach Learn Res 17(50):1–33

    MathSciNet  MATH  Google Scholar 

  3. Chen Z, Zhang K, Chan L, Schlkopf B (2014) Causal discovery via reproducing kernel hilbert space embeddings. Neural Comput 26(7):1484–1517 PMID: 24708374

    MathSciNet  Article  Google Scholar 

  4. Dhillon IS, Guan Y, Kulis B (2004) Kernel k-means: spectral clustering and normalized cuts. In: Proceedings of the tenth ACM SIGKDD international conference on knowledge discovery and data mining, KDD ’04, pp 551–556. ACM, New York

  5. Du N, Song L, Woo H, Zha H (2013) Uncover topic-sensitive information diffusion networks. In: Carvalho CM, Ravikumar P (eds), Proceedings of the sixteenth international conference on artificial intelligence and statistics, vol 31 of proceedings of machine learning research, pp 229–237. PMLR, Scottsdale, Arizona

  6. Du N, Song L, Yuan M, Smola AJ (2012) Learning networks of heterogeneous influence. In: Pereira F, Burges CJC, Bottou L, Weinberger KQ (eds) Advances in neural information processing systems 25. Curran Associates Inc, pp 2780–2788

  7. Easley D, Kleinberg J (2010) Networks, crowds, and markets: reasoning about a highly connected world. Cambridge University Press, New York

    Google Scholar 

  8. Fraley RC, Raftery AE (2002) Model-based clustering, discriminant analysis, and density estimation. J Am Stat Assoc 97:611–631

    MathSciNet  Article  MATH  Google Scholar 

  9. Gomez-Rodriguez M, Balduzzi D, Schölkopf B (2011) Uncovering the temporal dynamics of diffusion networks. In: Proceedings of the 28th international conference on international conference on machine learning, ICML’11, pp 561–568. Omnipress, USA

  10. Gomez-Rodriguez M, Leskovec J, Krause A (2012) Inferring networks of diffusion and influence. ACM Trans Knowl Discov Data 5(4):21:1–21:37

    Article  Google Scholar 

  11. Gomez-Rodriguez M, Leskovec J, Schölkopf B (2013) Structure and dynamics of information pathways in online media. In: Proceedings of the sixth ACM international conference on web search and data mining, WSDM ’13, pp 23–32. ACM, New York

  12. Gomez-Rodriguez M, Schölkopf B (2012) Influence maximization in continuous time diffusion networks. In: Proceedings of the 29th international conference on international conference on machine learning, ICML’12, pp 579–586. Omnipress, USA

  13. Gomez-Rodriguez M, Song L, Du N, Zha H, Schölkopf B (2016) Influence estimation and maximization in continuous-time diffusion networks. ACM Trans Inf Syst 34(2):9:1–9:33

    Article  Google Scholar 

  14. Goyal A, Bonchi F, Lakshmanan LV (2010) Learning influence probabilities in social networks. In: Proceedings of the third ACM international conference on web search and data mining, WSDM ’10, pp 241–250, ACM, New York

  15. Grabowicz PA, Ganguly N, Gummadi KP (2016) Distinguishing between topical and non-topical information diffusion mechanisms in social media. In: Proceedings of the 10th international conference on web and social media, pp 151–160

  16. Gretton A, Borgwardt KM, Rasch MJ, Schölkopf B, Smola A (2012) A kernel two-sample test. J Mach Learn Res 13(1):723–773

    MathSciNet  MATH  Google Scholar 

  17. Jamali M, Ester M (2010) A matrix factorization technique with trust propagation for recommendation in social networks. In: Proceedings of the fourth ACM conference on recommender systems, RecSys ’10, pp 135–142. ACM, New York

  18. Jegelka S, Gretton A, Schölkopf B, Sriperumbudur BK, von Luxburg U (2009) Generalized clustering via kernel embeddings. In: Mertsching B, Hund M, Aziz Z (eds) KI 2009: advances in artificial intelligence. Springer, Berlin, pp 144–152

    Google Scholar 

  19. Kempe D, Kleinberg J, Tardos E (2003) Maximizing the spread of influence through a social network. In: Proceedings of the ninth ACM SIGKDD international conference on knowledge discovery and data mining, KDD ’03, pp 137–146. ACM, New York

  20. Lei S, Maniu S, Mo L, Cheng R, Senellart P (2015) Online influence maximization. In: Proceedings of the 21th ACM SIGKDD international conference on knowledge discovery and data mining, KDD ’15, pp 645–654. ACM, New York

  21. McLachlan G, Peel D (2004) Finite mixture models. Wiley series in probability and statistics: applied probability and statistics. Wiley, London

  22. Muandet K, Fukumizu K, Sriperumbudur B, Schölkopf B (2017) Kernel mean embedding of distributions: a review and beyond. Foundations Trends Mach Learn 10(1–2):1–141

    Article  MATH  Google Scholar 

  23. Myers S, Leskovec J (2010) On the convexity of latent social network inference. In: Advances in neural information processing systems 23, pp 741–1749. Curran Associates, Inc

  24. Rahimi A, Recht B (2008) Random features for large-scale kernel machines. In: Advances in neural information processing systems 20, pp 1177–1184. Curran Associates, Inc

  25. Romero DM, Meeder B, Kleinberg J (2011) Differences in the mechanics of information diffusion across topics: Idioms, political hashtags, and complex contagion on twitter. In: Proceedings of the 20th international conference on world wide web, WWW ’11, pp 695–704. ACM, New York

  26. Rong Y, Zhu Q, Cheng H (2016) A model-free approach to infer the diffusion network from event cascade. In: Proceedings of the 25th ACM international on conference on information and knowledge management, CIKM ’16, pp 1653–1662. ACM, New York

  27. Rudin W (2017) Fourier analysis on groups. Dover books on mathematics. Dover Publications, NY

    Google Scholar 

  28. Saito K, Nakano R, Kimura M (2008) Prediction of information diffusion probabilities for independent cascade model. In: Proceedings of the 12th international conference on knowledge-based intelligent information and engineering systems, Part III, KES ’08, pp 67–75. Springer, Berlin

  29. Schölkopf B, Smola AJ (2001) Learning with Kernels: support vector machines, regularization, optimization, and beyond. MIT Press, Cambridge

    Google Scholar 

  30. Shirazi S, Harandi MT, Sanderson C, Alavi A, Lovell BC (2012) Clustering on grassmann manifolds via kernel embedding with application to action analysis. In: 2012 19th IEEE international conference on image processing, pp 781–784

  31. Smola A, Gretton A, Song L, Schölkopf B (2007) A hilbert space embedding for distributions. In: Hutter M, Servedio RA, Takimoto E (eds) Algorithmic learning theory. Springer, Berlin, pp 13–31

    Google Scholar 

  32. Song L, Fukumizu K, Gretton A (2013) Kernel embeddings of conditional distributions: a unified kernel framework for nonparametric inference in graphical models. IEEE Signal Process Mag 30(4):98–111

    Article  Google Scholar 

  33. Song L, Huang J, Smola A, Fukumizu K (2009) Hilbert space embeddings of conditional distributions with applications to dynamical systems. In: Proceedings of the 26th annual international conference on machine learning, ICML ’09, pp 961–968. ACM, New York

  34. Vaswani S, Lakshmanan V, Schmidt M (2015) Influence maximization with bandits. In: NIPS workshop on networks in the social and information sciences

  35. Villmann T, Biehl M, Hammer B, Verleysen M (2009) Similarity-based clustering: recent developments and biomedical applications, vol 5400. Springer, Berlin

    Google Scholar 

  36. Wang H, Shi X, Yeung D-Y (2017) Relational deep learning: a deep latent variable model for link prediction. In: AAAI conference on artificial intelligence, pp 2688–2694

  37. Watts D (2004) Six degrees: the science of a connected age. W. W. Norton

  38. Watts DJ, Dodds PS (2007) Influentials, networks, and public opinion formation. J Consum Res 34(4):441–458

    Article  Google Scholar 

  39. Wen Z, Kveton B, Valko M, Vaswani S (2017) Online influence maximization under independent cascade model with semi-bandit feedback. In: Guyon I, Luxburg UV, Bengio S, Wallach H, Fergus R, Vishwanathan S, Garnett R (eds) Advances in neural information processing systems 30. Curran Associates Inc, pp 3022–3032

Download references

Author information

Affiliations

Authors

Corresponding author

Correspondence to Bogdan Cautis.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Responsible editor: Jesse Davis, Elisa Fromont, Derek Greene and Bjørn Bringmann.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Hu, S., Cautis, B., Chen, Z. et al. Model-free inference of diffusion networks using RKHS embeddings. Data Min Knowl Disc 33, 499–525 (2019). https://doi.org/10.1007/s10618-018-00611-1

Download citation

Keywords

  • Diffusion networks
  • Edge inference
  • Clustering
  • Kernel embeddings