基于观点吸引势的舆情事件意见领袖发现算法构建  

Construction of Opinion Leader Discovery Algorithm Based on Opinion Attraction

作　　者：吴诗贤[1] 张必兰[2] 

机构地区：[1]重庆工商大学计算机科学与信息工程学院,重庆400067 [2]重庆工商大学图书馆,重庆400067

出　　处：《重庆工商大学学报（自然科学版）》2016年第4期86-90,共5页Journal of Chongqing Technology and Business University:Natural Science Edition

基　　金：教育部人文社会科学研究基金(13YJCZH197)

摘　　要：从物理学中场的思想出发,将论坛、博客等舆情事件的讨论平台看作一个舆情场,引入观点势函数描述舆情场内各网民间观点的相互吸引作用,基于各网民的时变观点状态建立有向的网民观点吸引网络,通过寻找具有高观点吸引势的网民节点实现舆情事件意见领袖的发现;该算法能有效避免传统的基于链接或回复关系的意见领袖发现算法中网民连接的依赖关系不明确的缺点。Inspired by the idea of physical fields, the discussion platform of each blog or public opinion event is regarded as an opinion field, an opinion attraction potential function is used to describe the interactive attraction from all net citizens in a public opinion field, the directional opinion attraction network of the net citizens is established based on the time-variant point state of the cyber citizens, and then the opinion leader is discovered by detecting the nodes of the net citizens with high attraction potential. This algorithm can effectively avoid the disadvantages of dependency relationship indeterminacy among the net citizens in traditional opinion leader discovery algorithm based on link or reply relation.

关 键 词：意见领袖发现 舆情场 观点吸引势 观点吸引网络 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]