Sentiment Analysis of China's Education Policy Online Opinion Based on Text Mining

With the in-depth development of China's education reform, the combination of educational public opinion and emotion analysis is conducive to the discovery of the evolution law of public sentiment and related factors, and provides ideas for the guidance of public opinion of relevant government departments. This paper proposes an analysis model of public opinion on the education policy network based on text mining, obtains network comment contents through Octopus crawler, preprocesses the data, and introduces the calculation formula of emotion heat, on which the evolution process of public opinion is divided into incubation period, outbreak period, spreading period and decline period. Combined with visualization and emotional heat maps of provinces and ages are constructed, so as to grasp the spatial-temporal evolution law of educational public opinions. Taking the postponement of the college entrance examination in 2020 as an example, the empirical analysis proves that the online public opinion analysis model of education policy can reflect the evolution law.