Research On Health Public Opinion Based On Big Data Text Mining Technology

With the advancement of social technology and the widespread use of the Internet,online public opinion has become an important form of public opinion and social development.Health is the foundation for promoting people's all-round development,and is an important symbol of national prosperity.Our country attaches great importance to the reform and development of national health and medical-health undertakings,and has already listed "healthy China" as a national strategy.Therefore,studying the content of online public opinion from the health perspective has practical significance.Based on the current analysis technology of online public opinion,this study propose two analysis methods: based on TF-IDF word co-occurrence association rule mining technology and adaptive weight adjustment method for sentiment index construction,respectively from the association rules and emotional tendencies.(1)TF-IDF word co-occurrence association rule mining technology: this study uses the TF-IDF method to extract text features,and then uses the word co-occurrence method to pre-process.Finally,conduct association rule mining.(2)Research on the sentiment tendency index using adaptive weight adjustment method: first of all,this study construct a sentiment dictionary and then determine the part of speech of words in the document according to the sentiment dictionary.Then construct the affective tendency index through the weight of the length of the article and the affective words,and finally analyze the healthy affective tendency index.This study selects the hot topics of health public opinion from 2007 to 2020 for empirical research.The results show the correlation between topics of public opinion and analysis the historical changes of Public opinion sentiment index.The whole health public opinion is mainly centered on the topic of "Medical Hygiene and Health".When analyzing the sentiment tendency index,it is found that the change of the overall sentiment tendency index is greatly affected by social health events.It further proves that this method is feasible in the analysis of public opinion health topics.