社交媒体舆论事件中用户大量的情感显现成为一种普遍现象，情感在舆论事件传播特别是网络传播的过程中发挥了重要的作用。选择长生生物“假疫苗”事件作为案例，采集事件相关微博223903条和微博账号166739个,进行三级细粒度情感分析。研究发现微博信息传播的两个主要渠道评论和转发中均存在情感相关性，但情感传递的关系是一种弱关系。对比微博评论，微博转发过程中情感相关性的特征更明显，且更细粒度的情感测量，这种相关性更显著。在同级别情感中“慌”“惧”和负向情感在传播的过程中相关性最明显。不同情感在社交媒体平台传播的过程中快速迭代，迅速达到稳定状态，即产生情感趋同的结果。

The frequent sentimental disclosure of users in public opinion events based on social media has become a common phenomenon, and emotion plays an important role in the process of disseminating public opinion events, especially on platform of the Internet. The "fake vaccine" event of Changsheng Biology was selected as the case, and 223903 micro-blog posts related to the event and 166739 micro-blog accounts were crawled for three level fine-grained sentiment analysis. It is found that there is a sentimental correlation between the two main channels of comment and retweet to disseminate the micro-blog posts, But the relationship of sentimental transmission is weak. Compared with Weibo commenting, the characteristics of emotional relevance in Weibo forwarding process are more obvious. And the more fine-grained sentiment is measured, the correlation is more significant. Investigating the same level of sentiments, panic, fear and negative emotions are most relevant in the transmission. In the same level of emotion, "flu", "fear" and negative emotion are most relevant in the process of communication. Different kinds of emotions quickly iterate on social media platform, and fleetly reach a stable state. In other words, the result of sentimental convergence is formed.