A Public Opinion Analysis Model Based On Microblog Social Relation Network

Public opinion is the aggregate of individual attitudes or beliefs held by the adult population in some area in a period. To dominate and lead the public opinion is one of important acts of maintaining social stability and Party ruling security. With the rapid expansion of information technology, Internet become the main platform of information releasing, exchanging and acquiring with a huge number of users. To manually go to each page of the information contained in the screening and analysis of statistics is unrealistic. Hence, instead of public opinion survey, public opinion analyzing and mining on the Internet information become more and more important.Microblog is a platform that allow people to share, acquire and broadcast information by multiple tools such as web, wap or desktop client. People post message under the 140 words count limit, and the other people who"follow"him will see the message immediately. Microblog is becoming a huge information source. As it is shown by the"Annual Report of Chinese Public Opinion 2010", Microblog has become the third part of the public opinion source of China.There are plenty of research about public opinion mining on traditional Internet media, such as news, bbs and blog, yet there is rare public opinion analysis model of the emerging Microblog. Hence, the public opinion analysis of Microblog information is becoming a hot topic. In this dissertation, we research on the public opinion analysis model of Microblog. We apply vertical search and text mining technology in the public opinion analysis, and do research on three model of public opinion: a hot topic detection model, a weight based secure model of public opinion issue, and a classification model of Microblog users. At the last two chapter of this dissertation we design and implement a practical public opinion analysis system on Microblog. In practical application, the models proved to be efficient.