摘要
文章针对微博事件相对于传统事件在传播过程中的新特征,提出了利用图论中二部图的理论来获取微博事件间的关联关系的新方法。文中给出了将微博事件和微博用户的关系转换为二部图网络的方法,并根据微博用户在微博事件中的角色特征,给出了微博用户的综合权重,由此来构造"微博事件——微博用户"加权二部图。通过对比多种二部图投影算法,提出了一种基于加权的一维投影算法,在保留二部图结构信息的基础上得出了微博事件间相互关联和影响的定量表示。最后通过实验验证了文章算法的合理性和正确性。
In this paper, we introduce a new method of using the theory of bipartite graph to analysis the correlation of microblogging event by comparing the new features in the process of transmission between microblogging event and traditional event. We give a method on how to convert the relationship between microblogging event and microblogging users to bipartite graph. According to the role of microblogging users in microblogging event, users’ general weight is given to construct the weighted bipartite graph of microblogging event-microblogging users. By comparing a variety of bipartite graph projection algorithm, this paper proposes a one-mode weighted projection algorithm, and comes to a quantitative representation of the correlation and inlfuences without losing the information of bipartite graph. Finally, a systematic experiment is conducted to verify the rationality and correctness of the proposed algorithm.
出处
《信息网络安全》
2014年第9期44-49,共6页
Netinfo Security
基金
国家高科技研究发展计划(863计划)[2012AA013002]
国家重点基础研究发展计划(973计划)[2013CB329601
2013CB329602]
关键词
微博事件
关联分析
加权二部图
二部图投影
microblogging event
correlation analysis
weighted bipartite graph
bipartite graph projection
