有效的社会媒体热点话题传播模型研究

南京大学学报(自然科学版) ›› 2015, Vol. 51 ›› Issue (1): 187–196.

有效的社会媒体热点话题传播模型研究

韩忠明*,张梦,李梦琪,莫倩,刘鹂

出版日期:2015-01-04 发布日期:2015-01-04
作者简介:(北京工商大学计算机与信息工程学院,北京,100048)
基金资助:
国家自然科学基金（61170112），北京市属高等学校科学技术与研究生教育创新工程建设项目（PXM2012_014213_000037）

Efficiently modeling information propagation of hot topics in social medias

Han Zhongming*, Zhang Meng, Li Mengqi, Mo Qian, Liu Li

Online:2015-01-04 Published:2015-01-04
About author:(Beijing Technology and Business University,Beijing,100048, China)

摘要/Abstract

摘要： 交互式社会媒体上的热点话题具有巨大的影响力,对热点话题进行建模和预测是一个非常重要但困难的问题.针对话题参与用户的特点进行了分析,构建了用户活跃度以及用户重入概率等模型的合理假设条件.根据话题发展模式和基于用户参与话题概率构建了单峰模型和多峰模型.分别基于两个不同数据集对模型进行了拟合和预测试验,试验结果表明,本文提出的模型在拟合与预测话题的发展趋势上的效果都优于SpikeM模型,尤其是对具有复杂波动发展模式的话题,本文提出的模型能很好地拟合与预测话题的波动.

Abstract: Hot topics on interactive social media websites enormously affect the incidence and development of the various events in both virtual and real world.Modeling and predicting information propagation process of hot topics are very important but difficult research problems. In this paper, characteristics of participants in hot topics are deeplyanalyzed. As a result, user activity degree, user popularity degreeanduser re-entrance probability are defined.The assumptions of traditionalinformation propagation models of hot topics are relaxedaccording to twofeatures in a hot topic: one user could participate the same topic many times and different usershave different activity degrees. According two types of propagation patterns of hot topics, two effective models are proposedbased on user participation probability. The first modelis used to model the single peak propagation patterntopicsand the second mode is used to model multi-peaks propagation pattern topics.Two datasets are selected from popular social media websites and comprehensive experiments are conducted.Two models proposed in this paper and SpikeM model are implemented for comparative study. The experimental results show that the models proposed in this paper can effectively simulate single peak propagation pattern and multi-peaks propagation pattern of hot topics. Especially, the models proposed in this paper outperform SpikeM model for fitting topics with complex rise-fall propagation patterns.Furthermore, the models canaccurately predict future propagation patterns of hot topics in real datasets.

韩忠明*,张梦,李梦琪,莫倩,刘鹂. 有效的社会媒体热点话题传播模型研究[J]. 南京大学学报(自然科学版), 2015, 51(1): 187–196.

Han Zhongming*, Zhang Meng, Li Mengqi, Mo Qian, Liu Li
. Efficiently modeling information propagation of hot topics in social medias[J]. Journal of Nanjing University(Natural Sciences), 2015, 51(1): 187–196.

参考文献

[1] Barabasi A L. The original of bursts and heavy tails in human dynamics. Nature, 2005, 435(7039):207~211.
[2] Yasuko M, Yasushi S, Prakash A B, et al. Rise and fall patterns of information diffusion:Model and implications.In:Proceedings of the 18th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Beijing, China:ACM Press, 2012:6~14.
[3] Anderson R M, May R M, Anderson B. Infectious diseases of humans: Dynamics and control. Oxford: Oxford University Press, 1992,268(23):3381.
[4] 倪顺江, 翁文国, 范维澄. 具有局部结构的增长无标度网络中传染病传播机制研究. 物理学报, 2009, 58(6):230~236.
[5] 张彦超, 刘云, 张海峰等. 基于在线社交网络的信息传播模型. 物理学报, 2010, 60(5):132~138.
[6] Yoshiaki M.Global stability of extended multi-group SIR epidemic models with patches through migration and cross patch infection.ActaMathematica Scientia.2013,33B(2):341~361.
[7] Hawkes A G, Oakes D. A cluster representation of a self-exciting process. Journal of Applied Probability, 1974, 2(11):493~503.
[8] Crane R, Sornette D. Robust dynamic classes revealed by measuring the response function of a social system. PNAS, 2008, 105(41):15649~15653.
[9] 徐致靖,祖正虎,许晴等. 基于自激点过程的恐怖活动建模研究.军事医学,2012,36(10):750~753
[10] 刁柏青,史铁林,杨叔子.自激过程与非齐次Poisson用于钢丝绳断丝计数过程的建模与预测.振动工程学报,1995,8(4):311~316
[11] Leskovec J, Backstrom L, Kleinberg J M. Meme-tracking and the dynamic of the news cycle.In:Proceedings of the 15th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York, USA:ACM Press, 2009:497~506.
[12] Yang J, Leskovec J. Modeling information diffusion in implicit networks.In:Proceedings of the 10thInternational Conference on Data Mining. New York, USA:ACM Press, 2010:599~608.
[13] McGlohon M, Leskovec J, FaloutsosC,et al. Finding patterns in blog shapes and blog evolution. In:Proceedings of International Conference on Weblogs and Social Media, Boulder, Colorado:ACM Press, 2007.
[14] 赵丽,袁睿翕,管晓宏. 博客网络中具有突发性的话题传播模型. 软件学报, 2009, 20(5):1384~1392.
[15] Leskovec J, McGlohon M, FaloutsosC,et al. Patterns of cascading behavior in large blog graphs. In: Proceedingsof the SIAM International Conference on Data Mining. New York,USA:ACM Press, 2007:551~556.
[16] Kumar R, Mahdian M, McGlohon M. Dynamics of conversations.In:Proceedings of the 16th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.New York,USA:ACM Press,2010: 553~562.
[17] Prakash B, Chakrabarti D, Faloutsos M, et al. Threshold conditions for arbitrary cascade models on arbitrary networks. In:Proceedings of the IEEE International Conference on Data Mining, Vancouver, Canada:IEEE Computer Society Press,2011: 537~546.
[18] 易成岐,鲍媛媛,薛一波等.新浪微博的大规模信息传播规律研究.计算机科学与探索.2013,7(6):551~561
[19] Yang J, Leskovec J. Patterns of temporal variation in online media.In:Proceedings of the 4th ACM International Conference on Web Search and Data Mining. New York, USA:ACM Press, 2011:177~186.
[20] 韩忠明, 陈妮, 乐嘉锦等. 面向热点话题时间序列的有效聚类算法研究. 计算机学报, 2012, 35(11):2337~2347.
[21] Hosking J R M, Wallis J R. Parameter and quantile estimation for the generalized Pareto distribution. Technometrics, 1987, 29(3):339~349.
[22] Michaela G, Jure L, Mary M, et al.Modelingblogdynamics.In:Proceedings of the 3th International AAAI Conference on Weblogs and Social Media,San Jose, California, USA:AAAI Press, 2009:26~33.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed