Comparative prediction of confirmed cases with COVID-19 pandemic by machine learning, deterministic and stochastic SIR models

TL;DR

This paper compares machine learning and SIR models (deterministic and stochastic) for predicting COVID-19 cases, providing insights into short-term pandemic trends and potential control measures.

Contribution

It introduces a combined approach using machine learning and SIR models with numerical methods for COVID-19 case prediction.

Findings

Predictions suggest pandemic may end soon in some countries.

Most countries will see the pandemic decline by early May.

Different models provide consistent short-term forecasts.

Abstract

In this paper, we propose a machine learning technics and SIR models (deterministic and stochastic cases) with numerical approximations to predict the number of cases infected with the COVID-19, for both in few days and the following three weeks. Like in [1] and based on the public data from [2], we estimate parameters and make predictions to help on how to find concrete actions to control the situation. Under optimistic estimation, the pandemic in some countries will end soon, while for most of the countries in the world, the hit of anti-pandemic will be no later than the beginning of May.