Internet public opinion chaotic prediction based on chaos theory and the improved radial basis function neural networks

The information of Internet public opinion was influenced by many netizens and net medias, the characteristics of this information are no rules, stochastic chaotic and are a nonlinear complex evolution system. The corresponding model is difficult to establish and effectively predict by the traditional methods based on statistical and machine learning. The characteristics of internet public opinion are chaotic, so the chaos theory was introduced to research firstly, the information of internet public opinion having chaotic characteristic was proved by the lyapunov index. Then the model to predict the development trend of internet public opinion was established by the phase space reconstruction theory. At last, the hybrid algorithm EMPSO-RBF which was based on EM algorithm and the RBF neural network optimized by the improved PSO algorithm was proposed to solve the model. The hybrid algorithm fully took the advantage of the EM clustering algorithm and improved PSO, the RBF neural network was improved by initializing the network structure in the early stage and optimizing the network parameters in the late. Firstly, the EM clustering algorithm was adopted to obtain the center value and variance and the radial basis function was improved with the combination of traditional Gauss model. Then the relevant network parameters were obtained by the improved PSO algorithm which was based on error optimizing constantly the network parameters. The model algorithm can be accurately simulated the time series of chaotic information by the experiments which were validated by different chaotic time series information, and it can better describe the development trend of different information of internet public opinion. The prediction results are made for government to monitor and guide the information of internet public opinion and benefit the social harmony and stability.