Research on Key Issues and Optimization Strategies for Emergency Response to Public Health Emergencies

Abstract

The handling of every major public health event is a test of risk early warning ability and national governance capacity and will form experience and lessons in social governance. In this paper, we use the improved Apriori algorithm to mine the classification of public health emergencies, construct public health emergency response indicators, and carry out feature screening and indicator system construction. On this basis, the selected areas are analyzed using the Em prediction model based on Markov chains and Bayesian networks. In this paper, A city is selected as the research object, and the Em prediction model is first tested for its performance. By comparing it with the RBF model and the ARIMA model, the prediction model has the best accuracy, while the RBF model has the lowest accuracy. Then, the Em model was used to cluster the derived social risks in City A, and the clustering centers of four risk indices were derived, which were 0.202, 0.358, 0.492, and 0.644, respectively. Secondly, the public health of City A was graded, and the risk grades were classified into four grades: mild, moderate, severe, and extra severe. Finally, according to the classification of the level of public health event characterization, the analysis can be seen that environmental factors and plains have a greater impact on the occurrence of public health events.