基于LeaderRank的意见领袖发现模型及其应用

Opinion Leader Discovery Model Based on LeaderRank and Its Application

[目的/意义]随着网络和社交媒体的发展,网络"意见领袖"在网络社区的信息传播和交流中发挥着越来越重要的作用,在社会生活的各个方面对网络民意产生巨大的影响。因此,识别网络"意见领袖",掌握其特征和规律成为了网络信息传播研究的重要方面。[方法/过程]在PageRank思想的基础上,利用文本的TF-IDF计算网络社区用户节点的连接强度,以此改进PageRank算法,提出一种LeaderRank方法用来评价网络社区用户节点的重要度,并结合其他指标及BP神经网络进行"意见领袖"的发现实验以及进一步的数据挖掘工作。[结果/结论]实验结果表明,该方法相较于神经网络具有更高的识别率,该方法可以灵活配合其他指标和方法使用,具有更好的适用性、扩展性和稳定性。

[Purpose/Significance]With the development of network and social media,network"opinion leaders"are playing a more and more important role in the information dissemination and communication of online communities,and have a huge impact on the network public opinion in all aspects of social life.Therefore,it has become an important aspect of network information dissemination to identify network"opinion leaders"and master their characteristics and laws.[Method/Process]On the basis of PageRank idea,this paper uses TF-IDF of text to calculate the connection strength of network community user nodes,so as to improve PageRank algorithm.A leaderRank method is proposed to evaluate the importance of network community user nodes.Combined with other indicators and BP neural network,the"opinion leader"discovery experiment and further data mining work are carried out.[Result/Conclusion]The experimental results show that this method has higher recognition rate than neural network,and the method can flexibly cooperate with other indexes and methods,and has better applicability and stability.