Combined prediction model of tuberculosis based on generalized regression neural network

Abstract

Tuberculosis is a major public health issue of global concern. This article analyzes the data of tuberculosis incidence in age groups between 0–60 years in China from 2004 to 2016. This article predicts the trend of tuberculosis incidence in ages between 0–60 years in 2020. For the same prediction method, a single model can only provide information from one perspective. In this article, the autoregressive moving average model and the gray prediction model are used to make predictions respectively. Then the generalized regression neural network is used to weight dynamically based on two models' prediction results. So prediction results of the neural network are more accurate, and provide a scientific basis for conducting corresponding prevention and control measures.