在线社会关系网络中消息流行度的建模与预测研究

项目名称： 在线社会关系网络中消息流行度的建模与预测研究

项目编号： No.61472400

项目类型： 面上项目

立项/批准年度： 2015

项目学科： 计算机科学学科

项目作者： 沈华伟

作者单位： 中国科学院计算技术研究所

项目金额： 80万元

中文摘要： 以社交网络和社会媒体为代表的在线社会关系网络中，信息传播是衔接信息内容、用户行为、网络结构的纽带，利用大量可感知、可计算的在线社会关系网络数据，分析影响信息传播的复杂关联因素，探索信息传播的固有模式和内在规律，并对信息传播态势进行准确预测，是提高对在线社会关系网络有效利用能力和科学管理水平的关键。然而，在线社会关系网络中，个体行为的不确定性、影响力的尺度多样性、影响信息传播因素的复杂关联，使得在线社会关系网络中消息流行度的建模和预测研究面临着严峻的科学技术挑战。本项目着眼于信息传播的微观机理，从个体行为、影响力和传播模型三个方面出发，研究在线社会关系网络中消息流行度的建模和预测，旨在揭示网络信息传播的内在规律，形成网络信息传播态势分析和预测的有效工具，并部署应用到相关业务单位，产生实在的社会和经济效益。

中文关键词： 社会计算；信息传播；流行度预测；社交网络分析；复杂网络

英文摘要： In online social networks, such as social networking sites and social media sites, information propagation is the tie among information content, individuals' behavior and the structure of underlying network. Nowadays, social data is increasingly available to investigate the interlinked factors in information propagation. This provides a unprecedented opportunity to discover the intrinsic patterns and fundamental laws, which emerge in or govern the information propagation. For a given message, predicting its popularity accurately is critical to the regulation and utilization of online social networks. However, the prediction of popularity is a very challenging task because of the spontaneous individual behavior, the multi-scale individual influence, and the interlinked factors influencing information propagation. In this proposal, we focus on the microscopic mechanism behind information propagation. For the purpose of popularity prediction, we first investigate the pattern of individual behavior, the spread of influence and the information propagation model. We aim to uncover the fundamental laws of information propagation and develop effective methods to predict the popularity of online content. The results and tools will be applied to companies or organizations with relevant interests.

英文关键词： Social Computing;Information Propagation;Popularity Prediction;Social Network Analysis;Complex Network