A method of network public opinion prediction based on the model of grey forecasting and hybrid fuzzy neural network
Xuegang Chen,Sheng Duan,Shanglin Li,Dong Liu,Hongbin Fan
DOI: https://doi.org/10.1007/s00521-023-08205-9
2023-02-01
Neural Computing and Applications
Abstract:Unexpected events occur frequently, and network public opinion prediction is one of the important research directions. Aiming at the problem that the current network public opinion prediction models mostly take improving the accuracy of the model as a breakthrough point, and lack the problem of exploring the law of public opinion communication, the study analyzes the current micro blog emergency propagation, focusing on introducing the implicit law of emotion vector, user browsing, and emergencies. At the same time, it studies the influencing factors causing the fluctuation of micro blog transmission of emergencies and selects the grey prediction model. The defects of the model are analyzed, and it has the constant increment problem and the lack of ability to deal with interference factors, and metabolic grey prediction model is used for the prediction of micro blog emergencies. At the same time, the concept of an incremental coefficient is introduced and the hybrid fuzzy neural network is adopted, the emotional knowledge is the key factor affecting the increment of grey prediction model. Use fuzzy neural network to analyze the micro blog emotional data generation, and obtain a mixed public opinion prediction model based on fuzzy neural network and grey prediction model. In the experimental process, the performance of the optimized prediction model is compared with that of the original prediction model, and a large number of data analyses prove that the optimized prediction model is effective. The experimental results show that the optimized prediction model has higher prediction accuracy.
computer science, artificial intelligence