面向非稳态霍克斯过程的格兰杰因果发现算法研究
Research on Algorithms of Granger Causality Discovery for Non-Stationary Hawkes Process
摘要: 发现非稳态霍克斯过程中潜在的格兰杰因果关系具有重要意义。现有的因果发现算法主要基于稳态性假设，无法适用于非稳态的情况。为此，文中提出了一种面向非稳态霍克斯过程的格兰杰因果发现算法：首先，建立非稳态霍克斯过程因果网络结构学习模型；然后，利用贪婪算法来发现某段霍克斯过程存在的模式；最后，利用基于极大似然估计的稀疏组套索(MLE-SGL)的方法发现模式对应的因果关系。在模拟数据上的实验效果验证了算法的正确性和有效性，并在交互式网络电视(IPTV)数据集上发现了一些不同模式及有趣的因果关系。
Abstract: The discovery of potential Granger Causality in the non-stationary Hawkes process is of great significance. Existing causal discovery algorithms are mainly based on the assumption of stationary situation and thus cannot be applied to non-stationary situations. For this reason, this paper proposes an algorithm named Granger Causality for Non-stationary Hawkes Process (GC- NOHP): First, the work establishes the non-stationary Hawkes process causal network structure learning model; Then, the greedy algorithm is implemented to discover the patterns that are hidden in a certain Hawkes process and an MLE-SGL-based method is utilized to discover the corresponding Granger causality of the patterns. The experimental results on synthetic data verified the correctness and effectiveness of the algorithm, and the algorithm also can find different patterns and some interesting causal relationships on the Interactive Internet TV (IPTV) data set.
文章引用：陈济斌, 蔡瑞初. 面向非稳态霍克斯过程的格兰杰因果发现算法研究[J]. 计算机科学与应用, 2021, 11(4): 821-831. https://doi.org/10.12677/CSA.2021.114084

参考文献

[1] Wei, C., Cai, R. and Hao, Z. (2019) Mining Hidden Non-Redundant Causal Relationships in Online Social Networks. Neural Computing and Applications, 32, 6913-6923. [Google Scholar] [CrossRef] 
[2] Cai, R., Zhang, Z. and Hao, Z. (2016) Understanding Social Causalities behind Human Action Sequences. IEEE Transactions on Neural Networks and Learning Systems, 28, 1801-1813. [Google Scholar] [CrossRef] 
[3] 蔡瑞初, 陈薇, 张坤, 郝志峰. 基于非时序观察数据的因果关系发现综述简[J]. 计算机学报, 2017, 40(6): 1470-1490.
[4] 陈薇, 蔡瑞初, 伍运金, 谢峰, 郝志峰. 基于多组典型相关变量的因果关系发现算法[J]. 计算机应用研究, 2021, 38(1): 53-56.
[5] Gurleen, K. and Sukhmani, A. (2011) Study of TV Viewership Patterns among Youngsters in Northern India. Zenith International Journal of Multidis-ciplinary Research, 1, 141-160.
[6] Levin, A.M., Levin, I.P. and Weller, J.A. (2005) A Multi-Attribute Analysis of Preferences for Online and Offline Shopping: Differences Across Products, Consumers, and Shopping Stages. Journal of Electronic Commerce Research, 6, 281-290.
[7] Daley, D.J. and Vere-Jones, D. (2007) An Introduction to the The-ory of Point Processes: Volume II: General Theory and Structure. Springer-Verlag, New York.
[8] Bacry, E., Delattre, S., Hoffmann, M. and Muzy, J.F. (2013) Some Limit Theorems for Hawkes Processes and Application to Financial Sta-tistics. Stochastic Processes and their Applications, 123, 2475-2499. [Google Scholar] [CrossRef] 
[9] Farajtabar, M. Gomez-Rodriguez, M., Wang, Y., Li, S., Zha, H. and Song, L. (2018) COEVOLVE: A Joint Point Process Model for Information Diffusion and Network Co-Evolution. Companion Proceedings of the Web Conference 2018, Lyon, April 2018, 473-477. [Google Scholar] [CrossRef] 
[10] 蔡瑞初, 谢泳, 陈薇, 曾艳, 郝志峰, 杜文俊. 面向社交媒体的直接因果网络发现算法[J]. 计算机应用研究, 2020, 37(9): 2689-2693.
[11] Reynaud-Bouret, P. and Schbath, S. (2010) Adaptive Estimation for Hawkes Processes: Application to Genome Analysis. The Annals of Statistics, 38, 2781-2822. [Google Scholar] [CrossRef] 
[12] Carstensen, L., Sandelin, A., Winther, O. and Hansen, N.R. (2010) Multivariate Hawkes Process Models of the Occurrence of Regulatory Elements. BMC Bioinformatics, 11, Article No. 456. [Google Scholar] [CrossRef] [PubMed]
[13] Xu, H., Farajtabar, M. and Zha, H. (2016) Learning Granger Causality for Hawkes Processes. 33th International Conference on Machine Learning, New York, 20-22 June 2016, 1717-1726.
[14] Zhou, K., Zha, H. and Song, L. (2013) Learning Social Infectivity in Sparse Low-Rank Net-works Using Multi-Dimensional Hawkes Processes.16th International Conference on Artificial Intelligence and Statistics, Scottsdale, 29 April-1 May 2013, 641-649.
[15] Eichler, M., Dahlhaus, R. and Dueck, J. (2017) Graphical Modeling for Multivariate Hawkes Processes with Nonparametric Link Functions. Journal of Time Series Analysis, 38, 225-242. [Google Scholar] [CrossRef] 
[16] Rizoiu, M.-A., Lee, Y., Mishra, S. and Xie, L. (2017) A Tutorial on Hawkes Processes for Events in Social Media. arXiv preprint arXiv:1708.06401.
[17] Hawkes, A.G. (1971) Spectra of Some Self-Exciting and Mutually Exciting Point Processes. Biometrika, 58, 83-90. [Google Scholar] [CrossRef] 
[18] Hallac, D., Vare, S. and Boyd, S. (2018) Toeplitz Inverse Covari-ance-Based Clustering of Multivariate Time Series Data. 27th International Joint Conference on Artificial Intelligence, Stockholm, 13-19 July 2018, 5254-5258. [Google Scholar] [CrossRef] 
[19] Møller, J. and Rasmussen, J.G. (2006) Approximate Simulation of Hawkes Processes. Methodology and Computing in Applied Probability, 8, 53-64. [Google Scholar] [CrossRef] 
[20] Luo, D., Xu, H., Zha, H., Du, J., Xie, R., Yang, X. and Zhang, W. (2014) You Are What You Watch and When You Watch: Inferring Household Structures from IPTV Viewing Data. IEEE Transactions on Broadcasting, 60, 61-72. [Google Scholar] [CrossRef]