摘要
随着社交网络的迅猛发展,大量的网民参与到话题讨论,积极推动着信息的传播。而意见领袖在信息传播中又起着决定性的作用。因此,如何在网络舆论中准确寻找意见领袖成为研究热点。针对传统的意见领袖研究方法没有将用户节点看作网络或仅仅把网络中的用户节点权重看成是回复次数叠加的问题,综合考虑回帖者的情感倾向,提出基于回帖者情感倾向性的意见领袖发现算法。实验表明,改进的意见领袖发现算法与传统的PR算法、UI-LR算法相比具有更高的识别精度。
With the rapid development of social networks,a large number of netizens participate in the discussion of topics,and actively promote the dissemination of information. And opinion leader in the dissemination of information plays a decisive role. Therefore,how to accurately find opinion leaders in the network of public opinion has become a research hotspot. For the traditional research method of opinion leaders does not regard the user node as a network or simply consider the weight of the user node in the network as the problem of the number of replies,taking into account the emotional tendency of the replies,we put forward the opinion leader discovery algorithm. Experiments show that the improved opinion leader discovery algorithm has higher recognition accuracy compared with the traditional PR algorithm and UI-LR algorithm.
出处
《计算机应用与软件》
2017年第10期66-70,96,共6页
Computer Applications and Software
基金
国家自然科学基金项目(81360230
81560296)
关键词
意见领袖
回帖者
情感倾向
舆论
Opinion leader Replies Emotional tendency Opinion
