计算机工程Issue(3)：7-14,8.DOI:10.3969/j.issn.1000-3428.2015.03.002
基于支持向量机的炒作微博识别方法
Hype Microblog Recognition Method Based on Support Vector Machine
摘要
Abstract
Microblog is not only a center or channel of mass media,but also involved in the formation,development and guidance of public opinions. The propagation of speculation microblog which is released from We-media,opinion leaders or some other users,causes microblog rumors,false hype,social mobilization and other problems. This paper analyzes the phenomenon of covert planning, mines the difference of the structure in communication networks and the incremental statistics of forwardings between the ordinary and the speculation. A novel algorithm for hype microblog recognition is proposed in this paper based on Support Vector Machine ( SVM) which uses the modularity peak spread and the average diameter of the shortest path in propagation network. The proposed method has advantages of less dependence on user profile information and is sensitive to the structure of propagation networks,and it has higher recognition accuracy.关键词
社交网络/炒作群体/炒作微博/社团模块度/网络直径/平均最短路径/支持向量机Key words
social network/hype group/hype microblog/community module degree/network diameter/average shortest path/Support Vector Machine( SVM)分类
信息技术与安全科学引用本文复制引用
董雨辰,刘琰,罗军勇,张进..基于支持向量机的炒作微博识别方法[J].计算机工程,2015,(3):7-14,8.基金项目
国家自然科学基金资助项目(61309007) （61309007）
国家“863”计划基金资助项目(2012AA012902) （2012AA012902）
国家科技支撑计划基金资助项目(2012BAH47B01)。 （2012BAH47B01）
