【摘要】 日本核废水排海事件在互联网引起了极大的反响，迅速放大扩散到社会多个方面形成了一次舆情事件，在一定程度上影响到了社会管理甚至社会的安定。由此可见及时捕捉网络舆情，分析其特点，相关职能部分据此采取化解防范措施，已经成为当前亟待解决的问题。针对这一需求，开发了一个基于TF-IDF和Word2Vec算法的舆情监测程序。首先对微博内容文本进行清洗和分词处理，后利用TF-IDF算法提取微博文本关键词；其次按照关键词权重排序并生成词云图；最后将单词转换为高维向量并可视化在二维平面上，为舆情监测提供决策依据。更多还原

【Abstract】 The incident of Japan’s nuclear wastewater discharge to the sea has caused great repercussions on the Internet. It has rapidly expanded and spread to many aspects of society, forming a public opinion event, which has affected social management and even social stability to a certain extent. It can be seen that timely capture of online public opinion, analysis of its-, and relevant functional departments taking preventive measures based on this have become urgent problems to be solved.A public opinion monitoring program based on TF-IDF and Word2Vec algorithm has been developed to meet this demand. Firstly,the Weibo content text is cleaned and segmented, and then the TF-IDF algorithm is used to extract Weibo text keywords. Secondly,the keywords are sorted by weight and a word cloud map is generated. Finally, the words are converted into high-dimensional-and visualized on a two-dimensional plane, providing decision-making basis for public opinion monitoring.更多还原