Application of SARIMA-SVM combination model in the prediction of gonorrhea incidence
Objective The aim is to explore the application of SARIMA model,SVM model and SARIMA-SVM combined model in the prediction of gonorrhea incidence,so as to provide scientific evidence for gonorrhea prevention and control.Methods The monthly incidence data of gonorrhea in China from 2010 to 2021 were selected,and the monthly incidence data of gonorrhea from 2010 to 2020 were used as training sets to establish SARIMA model,SVM model and SARIMA-SVM combination model,respectively.The monthly incidence data of gonorrhea in 2021 were used as the test set to evaluate the effect of the models.Results The optimal SARIMA model was SARIMA(0,1,1)(0,1,1)12.MAE,MAPE and RMSE fitted by SARIMA model,SVM model and SARIMA-SVM combination model were 0.032 4,5.547 0％,0.048 1,0.009 5,1.787 7％,0.018 8 and 0.005 3,1.015 6％,0.014 0,respectively.MAE,MAPE and RMSE predicted by SARIMA model,SVM model and SARIMA-SVM combination model were 0.042 7,5.776 6％,0.062 1％,0.011 2,1.484 3％,0.011 3 and 0.010 7,1.398 9％,0.022 9,respectively.Conclusions SARIMA-SVM combination model is better than SARIMA model and SVM model in fitting and predicting the incidence of gonorrhea,and the prediction accuracy is better.
NETL
NSTL
万方数据
