Skip to main content

Link prediction in evolving heterogeneous networks using the NARX neural networks


In this article, we propose a novel multivariate method for link prediction in evolving heterogeneous networks using a Nonlinear Autoregressive Neural Network with External Inputs (NARX). The proposed method combines (1) correlations between different link types; (2) the effects of different topological local and global similarity measures in different time periods; (3) nonlinear temporal evolution information; (4) the effects of the creation, preservation or removal of the links between the node pairs in consecutive time periods. We evaluate the performance of link prediction in terms of different AUC measures. Experiments on real networks demonstrate that the proposed multivariate method using NARX outperforms the previous temporal methods using univariate time series in different test cases.

This is a preview of subscription content, access via your institution.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10
Fig. 11
Fig. 12
Fig. 13
Fig. 14
Fig. 15
Fig. 16


  1. 1.

  2. 2. which is collected as part of a research project (110E027) supported by Technological Research Council of Turkey (TUBITAK).


  1. 1.

    Acar E, Dunlavy D, Kolda T (2009) Link prediction on evolving data using matrix and tensor factorizations. In: IEEE international conference on data mining workshops, 2009. ICDMW ’09, pp 262–269

  2. 2.

    Adamic LA, Adar E (2003) Friends and neighbors on the web. Soc Netw 25(3):211–230

    Article  Google Scholar 

  3. 3.

    Barabâsi AL, Jeong H, Néda Z, Ravasz E, Schubert A, Vicsek T (2002) Evolution of the social network of scientific collaborations. Phys A 311(3):590–614

    MathSciNet  Article  MATH  Google Scholar 

  4. 4.

    Bliss CA, Frank MR, Danforth CM, Dodds PS (2014) An evolutionary algorithm approach to link prediction in dynamic social networks. J Comput Sci 5(5):750–764

    MathSciNet  Article  Google Scholar 

  5. 5.

    Chiang M, Liou J, Wang J, Peng W, Shan M (2013) Exploring heterogeneous information networks and random walk with restart for academic search. Knowl Inf Syst 36(1):59–82

    Article  Google Scholar 

  6. 6.

    Davis D, Lichtenwalter R, Chawla NV (2011) Multi-relational link prediction in heterogeneous information networks. In: Proceedings of the 2011 international conference on advances in social networks analysis and mining, IEEE Computer Society, Washington, DC, USA, ASONAM ’11, pp 281–288

  7. 7.

    Ermis B, Acar E, Cemgil AT (2015) Link prediction in heterogeneous data via generalized coupled tensor factorization. Data Min Knowl Disc 29(1):203–236

    MathSciNet  Article  Google Scholar 

  8. 8.

    Foresse D, Hagan F (1997) GaussNewton approximation to Bayesian learning. In: International conference on neural networks, Houston, USA

  9. 9.

    Gaul W, Vincent D (2017) Evaluation of the evolution of relationships between topics over time. Adv Data Anal Classif 11(1):159–178

    MathSciNet  Article  Google Scholar 

  10. 10.

    Giordano G, Vitale MP (2011) On the use of external information in social network analysis. Adv Data Anal Classif 5(2):95–112

    MathSciNet  Article  MATH  Google Scholar 

  11. 11.

    Gunes I, Gunduz-Oguducu S, Cataltepe Z (2015) Link prediction using time series of neighborhood-based node similarity scores. Data Min Knowl Disc 30(1):147–180

    MathSciNet  Article  Google Scholar 

  12. 12.

    Huang Z, Lin DKJ (2009) The time-series link prediction problem with applications in communication surveillance. INFORMS J Comput 21(2):286–303

    Article  Google Scholar 

  13. 13.

    Huang W, Lai KK, Nakamori Y, Wang S, Yu L (2007) Neural networks in finance and economics forecasting. Int J Inf Technol Decis Mak 06(01):113–140

    Article  Google Scholar 

  14. 14.

    Li X, Du N, Li H, Li K, Gao J, Zhang A (2014) A deep learning approach to link prediction in dynamic networks, Ch. 33, pp 289–297

  15. 15.

    Liben-Nowell D, Kleinberg J (2003) The link prediction problem for social networks. In: Proceedings of the twelfth international conference on information and knowledge management. ACM, New York, NY, USA, CIKM ’03, pp 556–559

  16. 16.

    Lichtenwalter R, Chawla N (2014) Vertex collocation profiles: theory, computation, and results. SpringerPlus 3(1):116

    Article  Google Scholar 

  17. 17.

    Lichtenwalter RN, Lussier JT, Chawla NV (2010) New perspectives and methods in link prediction. In: Proceedings of the 16th ACM SIGKDD international conference on knowledge discovery and data mining. ACM, New York, NY, USA, KDD ’10, pp 243–252

  18. 18.

    Lpez-Yez I, Sheremetov L, Yez-Mrquez C (2014) A novel associative model for time series data mining. Pattern Recogn Lett 41:23–33

    Article  Google Scholar 

  19. 19.

    Lü L, Zhou T (2011) Link prediction in complex networks: a survey. Phys A 390(6):1150–1170

    Article  Google Scholar 

  20. 20.

    Menezes JMP, Barreto GA (2008) Long-term time series prediction with the NARX network: an empirical evaluation. Neurocomputing 71(16–18):3335–3343

    Article  Google Scholar 

  21. 21.

    Minervini P, d’Amato C, Fanizzi N (2016) Efficient energy-based embedding models for link prediction in knowledge graphs. J Intell Inf Syst 47(1):91–109

    Article  Google Scholar 

  22. 22.

    Mogotsi I, Manning CD, Raghavan P, Schütze H (2010) Introduction to information retrieval. Inf Retr 13(2):192–195

    Article  Google Scholar 

  23. 23.

    Moradabadi B, Meybodi MR (2016) Link prediction based on temporal similarity metrics using continuous action set learning automata. Phys A 460:361–373

    MathSciNet  Article  Google Scholar 

  24. 24.

    Munasinghe L, Ichise R (2012) Time score: a new feature for link prediction in social networks. IEICE Trans Inf Syst 95(3):821–828

    Article  Google Scholar 

  25. 25.

    Murata T, Moriyasu S (2007) Link prediction of social networks based on weighted proximity measures. In: IEEE/WIC/ACM international conference on web intelligence, pp 85–88

  26. 26.

    Newman M (2001) Clustering and preferential attachment in growing networks. Proc Nat Acad Sci 98(2):025–102

    Article  Google Scholar 

  27. 27.

    Nguyen D, Widrow B (1990) Improving the Learning speed of 2-layer neural networks by choosing initial values of the adaptive weights. In: 1990 international joint conference on neural networks (IJCNN), pp 21–26

  28. 28.

    Olden JD, Jackson DA (2002) Illuminating the black box: a randomization approach for understanding variable contributions in artificial neural networks. Ecol Model 154(1–2):135–150

    Article  Google Scholar 

  29. 29.

    Ozcan A, Gunduz-Oguducu S (2015) Multivariate temporal link prediction in evolving social networks. Int J Comput Inf Sci 16(3):24–34

    Google Scholar 

  30. 30.

    Qi X, Luo R, Fuller E, Luo R, Zhang CQ (2016) Signed quasi-clique merger: a new clustering method for signed networks with positive and negative edges. Int J Pattern Recognit Artif Intell 30(03):1650006

    MathSciNet  Article  Google Scholar 

  31. 31.

    Qiu D, Li H, Li Y (2014) Identification of active valuable nodes in temporal online social network with attributes. Int J Inf Technol Decis Mak 13(04):839–864

    Article  Google Scholar 

  32. 32.

    Rawashdeh M, Kim HN, Aljaam JM, El Saddik A (2013) Folksonomy link prediction based on a tripartite graph for tag recommendation. J Intell Inf Syst 40(2):307–325

    Article  Google Scholar 

  33. 33.

    Rümmele N, Ichise R, Werthner H (2015) Exploring Supervised methods for temporal link prediction in heterogeneous social networks. In: Proceedings of the 24th international conference on world wide web, Florence, Italy, WWW ’15 Companion, pp 1363–1368

  34. 34.

    Shi C, Li Y, Zhang Y, Sun Y, Yu P (2016) A survey of heterogeneous information network analysis. IEEE Trans Knowl Data Eng 99:1–46

    Google Scholar 

  35. 35.

    Soares PRS, PrudêNcio RBC (2013) Proximity measures for link prediction based on temporal events. Expert Syst Appl 40(16):6652–6660

    Article  Google Scholar 

  36. 36.

    Steinhaeuser K, Chawla NV (2010) Identifying and evaluating community structure in complex networks. Pattern Recogn Lett 31(5):413–421

    Article  Google Scholar 

  37. 37.

    Sun Y, Barber R, Gupta M, Aggarwal CC, Han J (2011) Co-author relationship prediction in heterogeneous bibliographic networks. In: Proceedings of the 2011 international conference on advances in social networks analysis and mining, IEEE Computer Society, Washington, DC, USA, ASONAM ’11, pp 121–128

  38. 38.

    Sun Q, Wang N, Zhou Y, Luo Z (2016) Identification of influential online social network users based on multi-features. Int J Pattern Recognit Artif Intell 30(06):1659015

    Article  Google Scholar 

  39. 39.

    Tang J, Zhang J, Yao L, Li J, Zhang L, Su Z (2008) Arnetminer: extraction and mining of academic social networks. In: Proceedings of the 14th ACM SIGKDD international conference on knowledge discovery and data mining, ACM, New York, NY, USA, KDD ’08, pp 990–998

  40. 40.

    Wang H, Song G (2014) Innovative NARX recurrent neural network model for ultra-thin shape memory alloy wire. Neurocomputing 134:289–295

    Article  Google Scholar 

  41. 41.

    Wetzker R, Zimmermann C, Bauckhage C (2008) Analyzing social bookmarking systems: A cookbook. In: Mining social data (MSoDa) workshop proceedings. ECAI 2008, pp 26–30

  42. 42.

    Wu Z, Lin Y, Wang J, Gregory S (2016) Link prediction with node clustering coefficient. Phys A 452:1–8

    Article  Google Scholar 

  43. 43.

    Xiang EW (2008) A survey on link prediction models for social network data

  44. 44.

    Yang Y, Chawla NV, Sun Y, Han J (2012) Link prediction in heterogeneous networks: influence and time matters. In: Proceeding of IEEE international conference on data mining

Download references

Author information



Corresponding author

Correspondence to Alper Ozcan.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Ozcan, A., Oguducu, S.G. Link prediction in evolving heterogeneous networks using the NARX neural networks. Knowl Inf Syst 55, 333–360 (2018).

Download citation


  • Heterogeneous social network analysis
  • Evolving networks
  • Node similarities
  • Link prediction
  • NARX