A Probe into the Influence of Major Infectious Diseases on the Grain Yield of Each Province based on E-SVR Method

Xiaoxing Tong, Liang Meng, Guo Yu

2022

Abstract

The influence of major infectious diseases to the grain yield of the province was investigated by establishing a new prediction method based on ε-support vector regression(ε-SVR). the train model was built from historical data, including the grain yield of Beijing, Tianjing etc affected by SARS-CoV in 2003, Guangzhou in 1961 affected by cholera, Xinjiang in 1986 affected by Hepatitis E. It is proved that γ in radial basis kernel function is 0.01, penalty coefficient C is 1.0e + 7, loss function P is 10, the average relative error of model fitting is 1.96%, and the decisive coefficient is 0.99. We predict the production data of Gansu, Shanxi and Guangdong affected by SARS in 2003 and that of Guangdong affected by break-bone fever in 1978.The average relative error was 3.27%. However, after removing the two factors of the proportion of the infected population and the proportion of dead population, the model was built again. The average relative error of model fitting was 1.97%, and the average relative error of prediction was 3.31%. It shows that the major infectious diseases only have a small impact on grain yield. This model provides a new method for regional grain yield prediction and national macro-control in the short term.

Paper Citation

in Harvard Style

Tong X., Meng L. and Yu G. (2022). A Probe into the Influence of Major Infectious Diseases on the Grain Yield of Each Province based on E-SVR Method. In Proceedings of the International Conference on Big Data Economy and Digital Management - Volume 1: BDEDM, ISBN 978-989-758-593-7, pages 895-899. DOI: 10.5220/0011356500003440

in Bibtex Style

@conference{bdedm22,
author={Xiaoxing Tong and Liang Meng and Guo Yu},
title={A Probe into the Influence of Major Infectious Diseases on the Grain Yield of Each Province based on E-SVR Method},
booktitle={Proceedings of the International Conference on Big Data Economy and Digital Management - Volume 1: BDEDM,},
year={2022},
pages={895-899},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0011356500003440},
isbn={978-989-758-593-7},
}

in EndNote Style

TY - CONF

JO - Proceedings of the International Conference on Big Data Economy and Digital Management - Volume 1: BDEDM,
TI - A Probe into the Influence of Major Infectious Diseases on the Grain Yield of Each Province based on E-SVR Method
SN - 978-989-758-593-7
AU - Tong X.
AU - Meng L.
AU - Yu G.
PY - 2022
SP - 895
EP - 899
DO - 10.5220/0011356500003440