Visualization Analysis and XGBoost-Based Prediction of COVID-19 Mortality

Zixin Chen1; Chujing Zhang2; Ngating Chun3

10.1109/ICIPCA59209.2023.10257874

2023

979-8-3503-1468-7

2023 IEEE International Conference on Image Processing and Computer Applications (ICIPCA)

70-74

11-13 Aug. 2023

Changchun, China

For the best predictive results of novel coronavirus infection and COVID-19 mortality, this research bases on the XGBoost machine learning algorithm. Through the research of data on related diseases, it not only helps to prevent the infection of COVID-19 effectively, but also gives advice to specific patients who need treatment without delay. The existing machine learning model uses the logistic regression algorithm to train the uneven data but gets an unsatisfying precision. This research improves the result by combining undersampling with XGBoost, with random forest, logistic regression, decision tree, and other models as comparisons. The prediction of COVID-19 mortality has the same accuracy of 91% before and after using the undersampling method, however, the AUC rises about 13% and finally reaches 92%• This research is available for the prediction of prevalence rate and death rate of COVID-19 in people who have basic diseases.

COVID-19 XGBoost machine learning

第一

EI

20234314933224

Decision trees ; Forecasting ; Learning algorithms ; Logistic regression ; Machine learning ; Patient treatment

Medicine and Pharmacology:461.6 ; Health Care:461.7 ; Artificial Intelligence:723.4 ; Machine Learning:723.4.2 ; Combinatorial Mathematics, Includes Graph Theory, Set Theory:921.4 ; Mathematical Statistics:922.2 ; Systems Science:961

IEEE

1.School of Biomedical Engineering, Southern University of Science and Technology, Shenzhen, Guangdong, China
2.School of International Pharmaceutical Business, China Pharmaceutical University, Nanjing, Jiangsu, China
3.School of Software Engineering, Shandong University, Jinan, Shandong, China

Zixin Chen,Chujing Zhang,Ngating Chun. Visualization Analysis and XGBoost-Based Prediction of COVID-19 Mortality[C],2023:70-74.