Network Public Opinion Prediction Based on Improved Particle Swarm Optimization and BP Neural Network

Abstract

The prediction of network public opinion can help to understand the development law and progress direction of network public opinion. It is helpful for relevant departments to deal with public opinion in a timely manner and has important value for the benign development of social public opinion. In order to provide a more effective forecasting method, this paper introduces the sample data and parameter setting of the prediction model on the basis of discussing the related technology of prediction model establishment. Finally, the IPSA-BP combined prediction model designed in this paper was compared with IPSA and BP model on the prediction results of five kinds of public opinion events. Experimental data showed that the error between the prediction results of IPSA-BP and the real value was about 2%. However, the error between the prediction results of IPSA and BP model and the real value is about 5%, so it is verified that the prediction ability of IPSA-BP has achieved the expected test effect.