随着Web2.0的发展，社交媒体已经成为人类进行信息交互的主要渠道。当突发公共事件时，民众通过网络迅速了解传播事件形成网络舆情。本研究以“新冠肺炎疫情”事件为例，通过分析比较微博和抖音两个社交媒体平台的网络舆情主题情感演化差异，为政府对应舆情危机提供建议。

本研究首先基于LDA模型进行主题聚类并基于KNN算法进行情感分类，再将数据综合计算并将计算结果可视化，最后对比分析微博和抖音在主题关键词集、主题热度演化过程和基于主题的情感强度演化过程三个方面的差异。

本研究发现在主题方面微博和抖音的主题关键词集和主题热度演化过程并不完全相同，且微博平台舆情主题比抖音平台舆情主题更加分散。在情感方面微博用户比抖音用户更加理性，更不容易被舆情裹挟；微博舆情消极情感态度影响程度弱，扩散速度快，持续时间长，消退速度慢；抖音舆情消极情感态度影响程度强，扩散速度慢，持续时间短，消退速度快。

With the development of Web2.0, social media has become the main channel for human information interaction. When public emergencies occur, people quickly understand the spread of events through the network to form online public opinion. This study takes the ' COVID-19 epidemic ' event as an example, and provides suggestions for the government to respond to the public opinion crisis by analyzing and comparing the emotional evolution differences of the network public opinion themes of MicroBlog and TikTok.

The research steps of this paper are as follows. Firstly, topic clustering is carried out based on LDA model, emotion classification is carried out based on KNN algorithm. Then, the data is comprehensively calculated and the calculation results are visualized. Finally, the differences between MicroBlog and TikTok in three aspects of topic keyword set, topic heat evolution process and topic-based emotional intensity evolution process are compared and analyzed.

The results of this study are as follows. In terms of topics, the keyword sets and topic heat evolution processes of MicroBlog and TikTok are not exactly the same, and the public opinion theme of MicroBlog platform is more dispersed than that of TikTok platform. In terms of emotion, MicroBlog users are more rational than TikTok users and less likely to be coerced by public opinion ; the negative emotional attitude of MicroBlog public opinion has weak influence, fast diffusion speed, long duration and slow regression speed; the negative emotional attitude of TikTok public opinion has a strong influence, with slow diffusion speed, short duration and fast regression speed.