With the development of the information age, Micro-blog, as the main carrier and platform of public opinion in universities, has gradually increased its influence. Focusing on the high fluctuation of public opinion in universities and other certain characters, this paper proposes a predictive model of public opinion heat trend based on Markov chain. In this model, it collects some index data and gives out time series of the public opinion heat value; then classifies the state space of public opinion heat trend in universities; thirdly, constructs state transition matrix of public opinion heat; and lastly, forecasts the trend variation interval of public opinion heat. Finally, the experimental results show that such model can effectively predict the trend of public opinion in universities and guide and manage them, providing a theoretical model basis for effect appraisal of public opinion in universities.
[1]
Yan Cheng-kua.
Research on Campus Network Based on Big Data
,
2014
.
[2]
Jumadi,et al.
Opinion mining on Twitter microblogging using Support Vector Machine: Public opinion about State Islamic University of Bandung
,
2016,
2016 4th International Conference on Cyber and IT Service Management.
[3]
Ming Xia Zhu,et al.
Study on the Operation Model of Colleges Network Learning Propaganda and Public Opinion Guidance Based on the Internet
,
2013
.
[4]
Weige Ji,et al.
The University Network Public Opinion Monitoring System Based on Multi-Agent
,
2015
.
[5]
Xue-Gang Chen,et al.
Research on trend prediction and evaluation of network public opinion
,
2017,
Concurr. Comput. Pract. Exp..