Forecasting United States COVID-19 Cases and Deaths Through Machine Learning
Somalwar, A.
Show abstract
COVID-19 has become a great national security problem for the United States and many other countries, where public policy and healthcare decisions are based on the several models for the prediction of the future deaths and cases of COVID-19. While the most commonly used models for COVID-19 include epidemiological models and Gaussian curve-fitting models, recent literature has indicated that these models could be improved by incorporating machine learning. However, within this research on potential machine learning models for COVID-19 forecasting, there has been a large emphasis on providing an array of different types of machine learning models rather than optimizing a single one. In this research, we suggest and optimize a linear machine learning model with a gradient-based optimizer for the prediction of future COVID-19 cases and deaths in the United States. We also suggest that a hybrid of a machine learning model for shorter range predictions and a Gaussian curve-fitting model or an epidemiological model for longer range predictions could greatly increase the accuracy of COVID-19 forecasting.
Matching journals
The top 7 journals account for 50% of the predicted probability mass.
Similar papers in this journal
Similar papers in this journal
Similar papers in this journal
- Estimation of COVID-19 recovery and decease periods in Canada using machine learning algorithms 95%
- Distribution of Incubation Period of COVID-19 in the Canadian Context: Modeling and Computational Study 94%
- A Recursive Bifurcation Model for Predicting the Peak of COVID-19 Virus Spread in United States and Germany 94%
Similar papers in this journal
- Predicting COVID-19 Pandemic in Saudi Arabia Using Modified Singular Spectrum Analysis 93%
- A full-scale agent-based model of Lombardy COVID-19 dynamics to explore social networks connectivity and vaccine impact on epidemic 92%
- The influence of SARS-CoV-2 variants of concern on national case fatality rates 91%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.