Predicting the incidence of varicella using an error backpropagation-based neural network model

Abstract

Objective To determine the applicability of an error back propagation (BP) -based neural network model in predicting the incidence of varicella in Panyu District of Guangzhou, and provide theoretical basis for the prevention and control of varicella. Methods Based on the incidence of varicella from 2006 to 2015 in Panyu District of Guangzhou, a BP-based neural network model was constructed to test the prediction accuracy of the model. The model was used to predict the incidence of varicella from 2016 to 2018. Results The prediction accuracy of the constructed model reached 99.867%. Using the model, the incidence of varicella in the region from 2016 to 2018 would be 153.36, 118.17, and 128.97 in 100, 000, respectively. Conclusion The BP-based neural network model offers high prediction accuracy and is suitable for the prediction and analysis of the incidence of varicella. Key words: Error back propagation algorithm neural network model; Varicella; Incidence