A Study of Using Machine Learning in Predicting COVID-19 Cases

Authors

  • Maleerat Maliyaem King Mongkut's University of Technology North Bangkok, Thailand https://orcid.org/0000-0003-1684-3865
  • Nguyen Minh Tuan King Mongkut's University of Technology North Bangkok, Thailand https://orcid.org/0000-0002-4035-1759
  • Demontray Lockhart King Mongkut's University of Technology North Bangkok, Thailand
  • Supattra Muenthong King Mongkut's University of Technology North Bangkok, Thailand

DOI:

https://doi.org/10.37256/ccds.3220221488

Keywords:

COVID-19, machine learning, algorithms, Random Forest, fatality

Abstract

With an unprecedented challenge to combat COVID-19, the prediction of confirmed cases is very important to ensure medical aid and healthy living conditions. In order to predict confirmed cases, the current study uses a dataset prepared by the White House Office of Science and Technology Policy which brought together companies and research to address questions concerning COVID-19. The importance of this was to identify factors that seem to affect the transmission rate of COVID-19. The focus of the current research, however, is to predict global cases of COVID-19. There have been many papers written about the prediction of confirmed cases and fatalities, but they failed to show promising results. Our research applies machine learning for predicting fatalities in the world using the COVID-19 Forecasting dataset from Kaggle. After trying several algorithms, our findings reveal that Logistic Regression, Decision Tree, KNeighbors, GaussianNB, and Random Forest algorithms provide the best predictions. Thus, the results show Random Forest as having the highest accuracy followed by Logistic Regression and Decision Tree. The results are promising opening up the door for further research.

Downloads

Published

2022-07-14

How to Cite

1.
Maliyaem M, Nguyen Minh Tuan, Lockhart D, Muenthong S. A Study of Using Machine Learning in Predicting COVID-19 Cases. Cloud Computing and Data Science [Internet]. 2022 Jul. 14 [cited 2024 Mar. 29];3(2):92-9. Available from: https://ojs.wiserpub.com/index.php/CCDS/article/view/1488