Abstract
User Identity Linkage (UIL) across social networks can be used to identify the accounts belonging to the same individual in multiple social networks, which is of great significance for user behavior analysis and network security supervision. By integrating deep learning into the process of UIL, the original intrinsic features of users can be retained more completely. However, the data distribution among different networks is different, and the existing UIL methods often ignore these differences, or only adjust a single step by domain adaptation mechanism, so it is difficult to ensure the accuracy of UIL. In this paper, we propose a new UIL model (called DualLink) which uses a dual domain adaptation mechanism to solve the problem of inconsistent data distribution. On one hand, a node embedding method based on adversarial domain adaptation is proposed to learn the node representation by considering the attributes, the topological structure and the difference between domains. On the other hand, a node matching method based on back-propagation domain adaptation is proposed to learn the suitable matching function by using back propagation neural network (BPNN). The feasibility and effectiveness of the key technology proposed in this paper are verified by experiments.
This research is supported by National Natural Science Foundation of China (62072084, 62072086), Fundamental Research Funds for the Central Universities (N2116008, N180716010).
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Li, X., et al.: RLINK: deep reinforcement learning for user identity linkage. World Wide Web 24(1), 85–103 (2021). https://doi.org/10.1007/s11280-020-00833-8
Zhou, T., Lim, E.-P., Lee, R.K.-W., Zhu, F., Cao, J.: Retrofitting embeddings for unsupervised user identity linkage. In: Lauw, H.W., Wong, R.C.-W., Ntoulas, A., Lim, E.-P., Ng, S.-K., Pan, S.J. (eds.) PAKDD 2020, Part I. LNCS (LNAI), vol. 12084, pp. 385–397. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-47426-3_30
Kou, Y., Li, X., Feng, S., Shen, D., Nie, T.: A cross-network user identification model based on two-phase expansion. In: Ni, W., Wang, X., Song, W., Li, Y. (eds.) WISA 2019. LNCS, vol. 11817, pp. 397–403. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-30952-7_39
Li, J., Li, Z., Lü, S.: Feature concatenation for adversarial domain adaptation. Expert Syst. Appl. 169, 114490 (2021)
Tzeng, E., Hoffman, J., Saenko, K., Darrell, T.: Adversarial discriminative domain adaptation. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 7167–7176 (2017)
Shen, X., Chung, F.L.: Deep network embedding for graph representation learning in signed networks. IEEE Trans. Cybern. 50(4), 1556–1568 (2018)
Ganin, Y., et al.: Domain-adversarial training of neural networks. J. Mach. Learn. Res. 17(1), 2030–2096 (2016)
Makhzani, A., Shlens, J., Jaitly, N., Goodfellow, I., Frey, B.: Adversarial autoencoders. arXiv preprint arXiv:1511.05644 (2015)
Ma, J., et al.: De-anonymizing social networks with random forest classifier. IEEE Access 6, 10139–10150 (2017)
Pedarsani, P., Figueiredo, D.R., Grossglauser, M.: A Bayesian method for matching two similar graphs without seeds. In: 2013 51st Annual Allerton Conference on Communication, Control, and Computing (Allerton), pp. 1598–1607. IEEE (2013)
Peled, O., Fire, M., Rokach, L., Elovici, Y.: Entity matching in online social networks. In: 2013 International Conference on Social Computing, pp. 339–344. IEEE (2013)
Levy, O., Goldberg, Y.: Neural word embedding as implicit matrix factorization. Adv. Neural Inf. Process. Syst. 27, 2177–2185 (2014)
Yu, M.: Entity linking on graph data. In: Proceedings of the 23rd International Conference on World Wide Web,pp. 21–26 (2014)
Fu, S., Wang, G., Xia, S., Liu, L.: Deep multi-granularity graph embedding for user identity linkage across social networks. Knowl.-Based Syst. 193, 105301 (2020)
Gretton, A., Borgwardt, K., Rasch, M.J., Scholkopf, B., Smola, A.J.: A kernel method for the two-sample problem. arXiv preprint arXiv:0805.2368 (2008)
Mu, X., Zhu, F., Lim, E.P., Xiao, J., Wang, J., Zhou, Z.H.: User identity linkage by latent user space modelling. In: Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 1775–1784 (2016)
Man, T., Shen, H., Liu, S., Jin, X., Cheng, X.: Predict anchor links across social networks via an embedding approach. IJCAI. 16, 1823–1829 (2016)
Zhou, F., Liu, L., Zhang, K., Trajcevski, G., Wu, J., Zhong, T.: DeepLink: a deep learning approach for user identity linkage. In: IEEE INFOCOM 2018-IEEE Conference on Computer Communications, pp. 1313–1321. IEEE (2018)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 Springer Nature Switzerland AG
About this paper
Cite this paper
Xu, B., Kou, Y., Wang, G., Shen, D., Nie, T. (2021). DualLink: Dual Domain Adaptation for User Identity Linkage Across Social Networks. In: Xing, C., Fu, X., Zhang, Y., Zhang, G., Borjigin, C. (eds) Web Information Systems and Applications. WISA 2021. Lecture Notes in Computer Science(), vol 12999. Springer, Cham. https://doi.org/10.1007/978-3-030-87571-8_2
Download citation
DOI: https://doi.org/10.1007/978-3-030-87571-8_2
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-87570-1
Online ISBN: 978-3-030-87571-8
eBook Packages: Computer ScienceComputer Science (R0)