The Research Of Micro-blog Public Opinion Based On Sentiment Analysis

Aiming at the rising social media----micro-blog, this article did research on the outbreak of the social public opinion’s prediction and management. The article made scientific classification and prediction on social public opinion on the micro-blog based on the relevant network technology and the research of social public opinion. The development of social public opinion can be divided into three parts by applying grey theory on the research of social public opinion. In view of the characteristics in the process of the information transmission, the article made judging standard in each part and put forward the improved model-----GM(1,1). In the second part of public opinion development, the software can get the emotion tendency of micro-blog social public opinion and judge the development stage of the micro-blog public opinion and the severity of it through word segmentation, clustering and support vector machine. In the fifth chapter of the article, the article made a demonstration through the Guangzhou Maoming’s PX event. Through the analysis of the event, the article concluded that the data was consistent with the develop tendency. In the third part of the social public opinion, the paper made clear judgments and tagged it. Through the experiment, it was demonstrated that the model’s feasibility on micro-blog public opinion analysis and prevention.