摘要
研究微博用户转发行为,预测微博转发概率,确定影响微博转发概率的因素,在热点挖掘、产品营销、舆情监控、谣言控制等方面有重要的现实意义.本文介绍了影响微博转发的用户特征,其中比较典型的有用户影响力、粉丝平均标签数、粉丝活跃度等特征.通过粉丝数-关注数算法、用户标签数算法、粉丝活跃度算法等分析了它们与微博转发之间的关联关系,并确定各个属性的阈值,这些阈值对微博转发预测起到了至关重要的作用.
Studying micro-blog user forwarding behavior,micro-blog forward probability,and determining the factors that affect the forwarding probability of micro-blog have important practical significance in the aspects of hot spot mining,product marketing,public opinion monitoring,rumor control and so on.This paper describes the user characteristics affecting micro-blog forward,among which the user influence,the number of fans,the average number of tags,fans active degree and so on are more typical.The micro-blog's relationship with the number of fans,the number of users,the number of tags and the active degree of the fans are analyzed,mico-blog's and the threshold value of each attribute is determined.These thresholds play a crucial role in the forward prediction of micro-blog.
出处
《南华大学学报(自然科学版)》
2016年第4期100-105,共6页
Journal of University of South China:Science and Technology
基金
湖南省哲学社会科学基金资助项目(14YBA335)
国家自然科学基金资助项目(61402220)
关键词
微博
用户特征
转发
预测
micro-blog
user characteristics
forward
forecast