An efficient topic propagation model based on self-exciting point process

被引：0
|
作者：
Han Z.-M. [1 ]
Zhang M. [1 ]
Tan X.-S. [1 ]
Duan D.-G. [1 ]
Si H.-L. [1 ]
机构：
[1] School of Computer and Information Engineering, Beijing Technology and Business University, Beijing
来源：
Jisuanji Xuebao/Chinese Journal of Computers | 2016年 / 39卷 / 04期
基金：
中国国家自然科学基金;
关键词：
Data mining; Hawkes process; Hot topics; Propagation model; Self-exciting point process; Social media; Social networks;
D O I：
10.11897/SP.J.1016.2016.00704
中图分类号：
学科分类号：
摘要：
Modeling propagation processes of hot topics on Internet has significant meaning and value. This paper focuses on modeling hot topics on Internet and proposes a topic propagation model (Self-Exciting Point Process Model,SEPPM) based on self-exciting Hawkes process. SEPPM models the propagation process of one topic as a random point process by using self-exciting effect of user participation. At the same time, SEPPM also takes external factors for propagation into account, thus puts forward a formal topic propagation model. To evaluate effectiveness of SEPPM, comprehensive simulation and empirical experiment are conducted. A simulation algorithm for SEPPM is proposed and the simulated results show that SEPPM can generate a variety of patterns of hot topics with different propagation characteristics. The experimental results on real datasets show that SEPPM can not only splendidly fit real topic propagation process, but also can effectively forecast spreading trend. © 2016, Science Press. All right reserved.
引用
收藏
页码：704 / 716
页数：12
相关论文
共 19 条
  • [1] Allan J., Papka R., Lavrenko V., Online new event detection and tracking, Proceedings of the 21st Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, pp. 37-45, (1998)
  • [2] He Y., Song L.-L., On discovery and tracking of hotspots in key areas, Journal of Southwest China Normal University (Natural Science Edition), 39, 7, pp. 67-72, (2014)
  • [3] Deng J.-W., Deng K.-Y., Li Y.-X., Topic detection and tracking method for Tibetan network, Advanced Materials Research, 860-863, pp. 2914-2917, (2014)
  • [4] Chen Z., Zhu K., Ying L., Detecting multiple information sources in networks under the SIR model, Proceedings of the 48th CISS Annual Conference, pp. 1-4, (2014)
  • [5] Wen S., Zhou W., Zhang J., Xiang Y., Modeling propagation dynamics of social network worms, Parallel and Distributed Systems, 24, 8, pp. 1633-1643, (2013)
  • [6] Kurihara S., The multi agent based information diffusion model for false rumor diffusion analysis, Proceedings of the 14th WWW Companion International Conference on World Wide Web Companion, pp. 1319-1320, (2014)
  • [7] Fang J., Dougherty E., Saraf P., Et al., Epidemiological modeling of news and rumors on Twitter, Proceedings of the 7th Workshop on Social Network Mining and Analysis, (2013)
  • [8] Stomakhin A., Short M., Bertozzi A., Reconstruction of missing data in social networks based on temporal patterns of Interactions, Inverse Problems, 27, 11, pp. 1-15, (2011)
  • [9] Zhao L., Yuan R.-X., Guan X.-H., Jia Q.-S., Bursty propagation model for incidental events in blog networks, Journal of Software, 20, 5, pp. 1384-1392, (2009)
  • [10] Gruhl D., Guha R., Liben-Nowell D., Information diffusion through blogspace, Proceedings of the 13th International Conference on World Wide Web, pp. 491-501, (2004)
12