COVID-19 Data Prediction and Visualization in Mainland China
Proceedings of SPIE - The International Society for Optical Engineering
; 12597, 2023.
Article
in English
| Scopus | ID: covidwho-20245120
ABSTRACT
Contemporarily, COVID-19 shows a sign of recurrence in Mainland China. To better understand the situation, this paper investigates the growth pattern of COVID-19 based on the research of past data through regression models. The proposed work collects the data on COVID-19 in Mainland China from January 21st, 2020, to April 30th, 2020, including confirmed, recovered, and death cases. Based on polynomial regression and support vector machine regressor, it predicts the further trend of COVID-19. The paper uses root mean squared error to evaluate the performance of both models and concludes that there is no best model due to the high frequency of daily changes. According to the analysis, support vector machine regressors fit the growth of COVID-19 confirmed case better than polynomial regression does. The best solution is to utilize different types of models to generate a range of prediction result. These results shed light on guiding further exploration of the growth of COVID-19. © 2023 SPIE.
COVID-19; Mainland China; Polynomial Regression; Support Vector Machine; Data visualization; Mean square error; Polynomials; Regression analysis; Support vector machines; Best model; Data prediction; Growth patterns; Mainland chinas; Performance; Regression modelling; Regression vectors; Root mean squared errors; Support vectors machine
Full text:
Available
Collection:
Databases of international organizations
Database:
Scopus
Type of study:
Experimental Studies
/
Prognostic study
Language:
English
Journal:
Proceedings of SPIE - The International Society for Optical Engineering
Year:
2023
Document Type:
Article
Similar
MEDLINE
...
LILACS
LIS