摘要
[目的/意义]基于理性行为理论分析微博用户转发行为,考虑了微博语义特征、数据形式、用户特征、用户交互等因素,探究了影响用户态度和主观规范的因素对用户转发行为的影响。[方法/过程]在分析微博特征、用户特征及用户转发行为的基础上,基于理性行为理论提出研究假设,采用主题建模的技术研究文本的语义信息,基于LDA进行微博文本与用户兴趣的概率主题分布的训练,基于二元Logistic回归模型对提出的影响因素完成数据验证。[结果/结论]态度因素中,接收用户活跃度、微博文本与用户兴趣的语义相似度对用户转发行为影响显著,微博数据形式对用户转发行为影响不显著;主观规范因素中,发布用户与接收用户交互度对用户转发行为影响显著,发布用户的影响力对用户转发行为影响不显著。研究成果在用户行为预测、微博传播分析等领域可借鉴使用。
[Purpose/Significance] In this paper, the reposting behavior of microblog users is analyzed based on the Theory of Reasoned Action, and the semantic information and the data form of a microblog, the user characteristics and the interaction between users are fully considered, and the impact of the factors that influence the user＇s attitude and subjective norm on user＇s reposting behavior is explored. [Method/Process] On the basis of analyzing the feature of microblog, user and the reposting behavior, the hypothesis is proposed based on the Theory of Reasoned Action. The method of topic modeling is utilized to get the semantic information of micro-blog text, and the topic probability distribution of micro-blog text and user interest are obtained based on the LDA. The hypothesis is verified based on binary lo-gistic regression with the crawled data. [Result/Conclusion]In the attitude aspect, the receiving user＇s activity and the semantic similarity of micro-blog text and user interest have a significant affection on the reposting behavior, while the data form is not significant. In the sub-jective norm aspect, the interaction degree between the receiving user and the releasing user has a significant affection on the reposting be-havior, while the impact of releasing user is not significant. The study may be used in behavior prediction, information dissemination and other fields as reference.
作者
陈姝
窦永香
张青杰
Chen Shu;Dou Yongxiang;Zhang Qingjie(School o f Economics and Management,Xidian University, Xi＇an 710126)
出处
《情报杂志》
CSSCI
北大核心
2017年第11期147-152,160,共7页
Journal of Intelligence
基金
陕西高校人文社会科学青年英才支持计划(编号:ER42015060001)资助
