Integrating Neural Networks and Big Data for Public Opinion Monitoring and Decision Support in Social Network

As society advances and the demand for social governance increases, the collection of public opinions and decision-making support have gradually become the bottleneck in enhancing governance. In this paper, we put forward a model of public opinion monitoring and decision support based on fusion of graph neural network (GNN) and big data. The model innovatively applies graph neural networks for modeling the public opinion propagation in social network on the basis of traditional sentiment analysis and public opinion mining methods, and can capture the complex emotional correlation and public opinion evolution law hidden in the social network. What is more, combined with the node and edge based feature extraction method, is used to optimize the spatiotemporal dynamics of public opinion propagation, which aims to improve the accuracy and timeliness of prediction. The experimental results demonstrate that the model has a higher accuracy and a strong real-time response ability compared with the existing methods in public opinions monitoring and decision-making support tasks and can provide a more intelligent and accurate decision-making basis for social governance.