社会网络中信息的扩散机理及其定量建模

发布时间：2018-04-20 18:45

本文选题：社会网络 + 信息扩散模型　；参考：《中国管理科学》2017年12期

【摘要】：在自媒体时代,信息来源更多、扩散速度更快、扩散范围更广,这使得网络舆论的监控和管理更加困难。为提高网络舆情的管控能力,需深入了解网络中信息的扩散过程及重要特征。本文首先从微观角度分析信息的扩散机理,考虑网络的结构特点以及舆情监控的时效性,引入与用户是否在线相对应的节点状态,将传统的独立级联模型扩展为基于离散时间的双概率独立级联扩散模型。接着,本文从宏观角度对信息的扩散过程进行分析并定量表示,结合信息自身质量和用户网络特征两个客观要素,并考虑外部平台的影响,进而建立有关事件的动态扩散方程。与以往研究只利用模型模拟或纯粹用数值拟合相比,本文实验给出了社会网络中的信息扩散过程的一个更精准的刻画。实验结果展现了信息传播的多种形态,同时发现用户规模越大、用户关联越紧密以及信息质量越高时,信息爆发所需的时间越短。这些发现有利于预测信息扩散的趋势,同时为舆情管控的时效性和网络用户群体提供参考。
[Abstract]:In the era of self-media, more information sources, faster diffusion speed and wider spread range, which makes the monitoring and management of network public opinion more difficult. In order to improve the control ability of network public opinion, it is necessary to deeply understand the diffusion process and important characteristics of information in the network. In this paper, firstly, the mechanism of information diffusion is analyzed from the micro point of view, considering the structural characteristics of the network and the timeliness of the monitoring of public opinion, the node state corresponding to whether the user is online is introduced. The traditional independent cascade model is extended to a discrete time based double probabilistic independent cascade diffusion model. Then, this paper analyzes and quantifies the diffusion process of information from a macro perspective, combines the two objective elements of information quality and user network characteristics, and considers the influence of external platforms, and then establishes the dynamic diffusion equation of relevant events. Compared with the previous studies, we present a more accurate description of the information diffusion process in social networks. The experimental results show a variety of forms of information dissemination. At the same time, it is found that the larger the user size, the closer the user association and the higher the information quality, the shorter the time required for the information burst. These findings are helpful to predict the trend of information diffusion and provide a reference for the timeliness of public opinion control and network users.
【作者单位】：中国人民大学信息学院;
【基金】：国家自然科学基金资助项目(71531012,71271211) 北京市自然科学基金资助项目(4172032) 京东商城电子商务研究项目(413313012) 中国人民大学科学研究基金项目(10XNI029)
【分类号】：G206;TP393.09

【相似文献】