Abstract
Influence Maximization (IM) aims to maximize the number of people that become aware of a product by finding the “best” set of “seed” users to initiate the product advertisement. Unlike most prior arts on the static networks containing fixed number of users, we study the evolving IM in more realistic evolving networks with temporally growing topology. The task of evolving IM, however, is far more challenging over static cases in the sense that the seed selection should consider its impact on future users who will join network during influence diffusion and the probabilities that users influence one another also evolve over time.
We address the challenges brought by network evolution through EIM, a newly proposed bandit-based framework that alternates between seed nodes selection and knowledge (i.e., nodes’ growing speed and evolving activation probabilities) learning during network evolution. Remarkably, the EIM framework involves three novel components to handle the uncertainties brought by evolution: (1) A fully adaptive particle learning of nodes’ growing speed for accurately estimating future influenced size, with real growing behaviors delineated by a set of weighted particles. (2) A bandit-based refining method with growing arms to cope with the evolving activation probabilities via growing edges from previous influence diffusion feedbacks. (3) Evo-IMM, an evolving seed selection algorithm, which leverages the Influence Maximization via Martingale (IMM) framework, with the objective to maximize the influence spread to highly attractive users during evolution. Theoretically, the EIM framework returns a regret bound that provably maintains its sublinearity with respect to the growing network size. Empirically, the effectiveness of the EIM framework is also validated with three notable million-scale evolving network datasets possessing complete social relationships and nodes’ joining time. The results confirm the superiority of the EIM framework in terms of an up to 50% larger influenced size over four static baselines.
Supplemental Material
Available for Download
Supplemental movie, appendix, image and software files for, Evolving Influence Maximization in Evolving Networks
- X. Wu, L. Fu, J. Meng, and X. Wang. 2019. Maximizing influence diffusion over evolving social networks. In Proceedings of the ACM SocialSense. 6--11.Google Scholar
- Q. Wu, Z. Li, H. Wang, W. Chen, and H. Wang. 2019. Factorization bandits for online influence maximization. In Proceedings of the ACM SIGKDD. 636--646.Google Scholar
- Y. Lin, W. Chen, and J. C. Lui. 2017. Boosting information spread: An algorithmic approach. In Proceedings of the IEEE ICDE. 883--894.Google Scholar
- K. Han, C. Xu, F. Gui, S. Tang, H. Huang, and J. Luo. 2018. Discount allocation for revenue maximization in online social networks. In Proceedings of the ACM MobiHoc. 121--130.Google Scholar
- G. Tong, W. Wu, S. Tang, and D. Du. 2016. Adaptive influence maximization in dynamic social networks. IEEE/ACM Trans. Netw. 25, 1 (2016), 112--125.Google Scholar
- J. Yuan and S. J. Tang. 2017. Adaptive discount allocation in social networks. In Proceedings of the ACM MobiHoc. 1--10.Google Scholar
- X. Wu, L. Fu, Y. Yao, X. Fu, X. Wang, and G. Chen. 2018. GLP: A novel framework for group-level location promotion in geo-social networks. IEEE/ACM Trans. Netw. 26, 6 (2018), 2870--2883.Google Scholar
Digital Library
- Y. Tang, Y. Shi, and X. Xiao. 2015. Influence maximization in near-linear time: A martingale approach. In Proceedings of the ACM SIGMOD. 1539--1554.Google Scholar
- H. Zhuang, Y. Sun, J. Tang, J. Zhang, and X. Sun. 2013. Influence maximization in dynamic social networks. In Proceedings of the IEEE ICDM. 1313--1318.Google Scholar
- S. Lei, S. Maniu, L. Mo, R. Cheng, and P. Senellart. 2015. Online influence maximization. In Proceedings of the ACM SIGKDD. 645--654.Google Scholar
- T. K. Shih, W. Gunarathne, A. Ochirbat, and H. M. Su. 2018. Grouping peers based on complementary degree and social relationship using genetic algorithm. ACM Trans. Internet Technol. 19, 1 (2018), 2.Google Scholar
Digital Library
- D. Kempe, J. Kleinberg, and É. Tardos. 2003. Maximizing the spread of influence through a social network. In Proceedings of the ACM SIGKDD. 137--146.Google Scholar
Digital Library
- H. T. Nguyen, T. P. Nguyen, T. N. Vu, and T. N. Dinh. 2017. Outward influence and cascade size estimation in billion-scale networks. In Proceedings of the ACM Sigmetrics. 1, 1 (2017), 1--30.Google Scholar
Digital Library
- Y. Tang, X. Xiao, and Y. Shi. 2014. Influence maximization: Near-optimal time complexity meets practical efficiency. In Proceedings of the ACM SIGMOD. 75--86.Google Scholar
- E. Cohen, D. Delling, T. Pajor, and R. F. Werneck. 2014. Sketch-based influence maximization and computation: Scaling up with guarantees. In Proceedings of the ACM CIKM. 629--638.Google Scholar
- Wechat. 2017. Retrieved from http://www.wechat.com/en/.Google Scholar
- C. Zang, P. Cui, and C. Faloutsos. 2016. Beyond sigmoids: The NetTide model for social network growth, and its applications. In Proceedings of the ACM SIGKDD. 2015--2024.Google Scholar
- H. Li, S. Bhowmick, J. Cui, and J. Ma. 2017. Time is what prevents everything from happening at once: Propagation time-conscious influence maximization. arXiv preprint arXiv:1705.10977 (2017).Google Scholar
- J. Iribarren and E. Moro. 2009. Impact of human activity patterns on the dynamics of information diffusion. Phys. Rev. Lett. 103, 3 (2009), 038702.Google Scholar
Cross Ref
- N. T. Gayraud, E. Pitoura, and P. Tsaparas. 2015. Diffusion maximization in evolving social networks. In Proceedings of the ACM COSN. 125--135.Google Scholar
- Z. Wen, B. Kveton, M. Valko, and S. Vaswani. 2017. Online influence maximization under independent cascade model with semi-bandit feedback. In Proceedings of the NIPS. 3022--3032.Google Scholar
- Y. Bao, X. Wang, Z. Wang, C. Wu, and F. C. Lau. 2016. Online influence maximization in non-stationary social networks. In Proceedings of the IEEE/ACM IWQoS. 1--6.Google Scholar
- Y. Yang, X. Mao, J. Pei, and X. He. 2016. Continuous influence maximization: What discounts should we offer to social network users? In Proceedings of the ACM SIGMOD. 727--741.Google Scholar
- J. Jankowski, R. Michalski, and P. Kazienko. 2013. Compensatory seeding in networks with varying avaliability of nodes. In Proceedings of the IEEE/ACM ASONAM. 1242--1249.Google Scholar
- R. Michalski, T. Kajdanowicz, P. Bródka, and P. Kazienko. 2014. Seed selection for spread of influence in social networks: Temporal vs. static approach. New Gen. Comput. 32, 3–4 (2014), 213--235.Google Scholar
Cross Ref
- L. Sun, W. Huang, P. S. Yu, and W. Chen. 2018. Multi-round influence maximization. In Proceedings of the ACM SIGKDD. 2249--2258.Google Scholar
- S. C. Lin, S. D., Lin, and M. S. Chen. 2015. A learning-based framework to handle multi-round multi-party influence maximization on social networks. In Proceedings of the ACM SIGKDD. 695--704.Google Scholar
- A. L. Barabási and R. Albert. 1999. Emergence of scaling in random networks. Science, 286, 5439 (1999), 509--512.Google Scholar
Cross Ref
- B. Bollobás and O. Riordan. 2004. The diameter of a scale-free random graph. In Combinatorica, 24, 1 (2004), 5--34.Google Scholar
- C. Zang, P. Cui, C. Faloutsos, and W. Zhu. 2017. Long short memory process: Modeling growth dynamics of microscopic social connectivity. In Proceedings of the ACM SIGKDD. 565--574.Google Scholar
- S. A. Myers and J. Leskovec. 2010. On the convexity of latent social network inference. In Proceedings of the NIPS. 1741--1749.Google Scholar
- M. Gomez-Rodriguez, J. Leskovec, and A. Krause. 2012. Inferring networks of diffusion and influence. ACM Trans. Knowl. Discov. Data 5, 4 (2012), 21.Google Scholar
Digital Library
- Z. Shen, W. X. Wang, Y. Fan, Z. Di, and Y. C. Lai. 2014. Reconstructing propagation networks with natural diversity and identifying hidden sources. Nat. Commun. 5, 1 (2014), 1--10.Google Scholar
Cross Ref
- S. Lamprier, S. Bourigault, and P. Gallinari. 2015. Extracting diffusion channels from real-world social data: A delay-agnostic learning of transmission probabilities. In Proceedings of the IEEE/ACM ASONAM. 178--185.Google Scholar
- M. Salathé, M. Kazandjieva, J. W. Lee, P. Levis, M. W. Feldman, and J. H. Jones. 2010. A high-resolution human contact network for infectious disease transmission. Proc. Nat. Acad. Sci. 107, 51 (2010), 22020--22025.Google Scholar
Cross Ref
- C. Zeng, Q. Wang, S. Mokhtari, and T. Li. 2016. Online context-aware recommendation with time varying multi-armed bandit. In Proceedings of the ACM SIGKDD. 2025--2034.Google Scholar
- A. C. Harvey. 1990. Forecasting, Structural Time Series Models and the Kalman Filter. Cambridge University Press.Google Scholar
- W. Chen, Y. Wang, and Y. Yuan. 2013. Combinatorial multi-armed bandit: General framework and applications. In Proceedings of the ACM ICML. 151--159.Google Scholar
- W. Chen, C. Wang, and Y. Wang. 2010. Scalable influence maximization for prevalent viral marketing in large-scale social networks. In Proceedings of the ACM SIGKDD. 1029--1038.Google Scholar
- P. Rigollet and J. C. Hütter. 2019. High dimensional statistics: Lecture notes. Retrieved from http://www-math.mit.edu/rigollet/PDFs/RigNotes17.pdf.Google Scholar
- Microsoft Academic Graph. 2016. Retrieved from https://www.microsoft.com/en-us/research/project/microsoft-academic-graph/.Google Scholar
- Xudong Wu, Luoyi Fu, Zixin Zhang, Huang Long, Jingfan Meng, Xinbing Wang, and Guihai Chen. 2019. Supplementary Material. Retrieved from http://www.cs.sjtu.edu.cn/fu-ly/paper/EIMfull_ACM_Tran_supp.pdf.Google Scholar
Index Terms
Evolving Influence Maximization in Evolving Networks
Recommendations
Maximizing Influence Diffusion over Evolving Social Networks
SocialSense'19: Proceedings of the Fourth International Workshop on Social SensingInfluence diffusion in social networks has been intensively studied over last two decades. Most prior arts assume that the underlying network structure is static, remaining fixed during the influence diffusion process. However, many real networks are ...
Evolving neural networks
GECCO '08: Proceedings of the 10th annual conference companion on Genetic and evolutionary computationNeuroevolution, i.e. evolution of artificial neural networks, has recently emerged as a powerful technique for solving challenging reinforcement learning problems. Compared to traditional (e.g. value-function based) methods, neuroevolution is especially ...
Evolving Social Networks via Friend Recommendations
SITIS '15: Proceedings of the 2015 11th International Conference on Signal-Image Technology & Internet-Based Systems (SITIS)A social network grows over a period of time withthe formation of new connections and relations. In recent yearswe have witnessed a massive growth of online social networkslike Facebook, Twitter etc. So it has become a problem ofextreme importance to ...






Comments