Journal of Frontiers of Computer Science and Technology ›› 2016, Vol. 10 ›› Issue (11): 1524-1531.DOI: 10.3778/j.issn.1673-9418.1509083

Previous Articles     Next Articles

Prediction of “Forwarding Whom” Behavior in Information Diffusion

BAO Peng1,2, SHEN Huawei1+, CHENG Xueqi1   

  1. 1. Institute of Computing Technology, Chinese Academy of Sciences, Beijing 100190, China
    2. School of Software Engineering, Beijing Jiaotong University, Beijing 100044, China
  • Online:2016-11-01 Published:2016-11-04


鲍  鹏1,2,沈华伟1+,程学旗1   

  1. 1. 中国科学院 计算技术研究所,北京 100190
    2. 北京交通大学 软件学院,北京 100044

Abstract: On online social networks, follow-ship network among users underlies the diffusion dynamics of messages; meanwhile, the structure of underlying social network determines the visibility of messages and forwarding activities in the diffusion process. Taking SinaWeibo as an example, this paper focuses on multiple exposure phenomena in information diffusion, and investigates the patterns and regularities of users?? forwarding behavior among multiple exposures combined with the structure of follow-ship network. This paper analyzes the “forwarding whom” problem of users among multiple exposures in information diffusion, aiming to model and predict the forwarding behavior of individuals, combining content features, network structure, temporal and historical information. The experimental results demonstrate that the new method achieves a high accuracy of 91.3%.

Key words: online social network, information diffusion, multiple exposures, forwarding whom

摘要: 在线社会关系网络中,用户之间的关注关系网络承载着上层的信息传播,关注关系网络的结构影响着消息的可见度,并影响着信息传播过程的转发选择。以新浪微博为例,围绕信息传播中的多次暴露现象展开研究,结合用户关注关系网络的结构,探索信息传播过程中多次暴露情形下用户转发选择行为的模式和规律。针对信息传播中用户在多个暴露源下的转发选择预测问题,融合消息内容、网络结构、时序和交互历史等多方面因素,建模和预测用户转发选择。实验结果表明,新方法的预测准确率高达91.3%。

关键词: 在线社会关系网络, 信息传播, 多次暴露, 转发选择