Authors :
Md Mahbubur Rahman; Farhana Tazmim Pinki
Volume/Issue :
Volume 8 - 2023, Issue 6 - June
Google Scholar :
https://bit.ly/3TmGbDi
Scribd :
https://tinyurl.com/23av8325
DOI :
https://doi.org/10.5281/zenodo.8098699
Abstract :
COVID-19 is a human-to-human transmissible
virus responsible for damage to the human body, and
people died all over the world. Bangladesh was affected
by COVID-19 on March 8th, 2020. During the pandemic,
people and the government struggled to prevent
transmission due to an inadequate supply of vaccines and
healthcare equipment. Therefore, it is essential to
understand the upcoming infected cases for several days.
That may help people and the government make pre-
decision before the pandemic to save live. In this paper,
we proposed a COVID-19 short-term forecasting model
using Linear Regression (LR), Least Absolute Shrinkage
and Selection Operation (LASSO) Regression, and
Support Vector Regression (SVR) to predict the next
seven days of COVID-19 infected cases in Bangladesh
during the pandemic situation. Here we considered data
from 8th May 2021 to 21st July 2021. We analyzed
different past data volumes for the model to understand
the impact of past data in the model. The result reveals
that Support Vector Regression (SVR) performance was
better than LR and LASSO in all aspects with high
accuracy. The performance also indicated that the high
volume of past data helps to increase prediction accuracy.
Keywords :
COVID-19, Short-term Forecast, New infected cases, Bangladesh, Supervised Machine Learning, LR, LASSO, SVR.
COVID-19 is a human-to-human transmissible
virus responsible for damage to the human body, and
people died all over the world. Bangladesh was affected
by COVID-19 on March 8th, 2020. During the pandemic,
people and the government struggled to prevent
transmission due to an inadequate supply of vaccines and
healthcare equipment. Therefore, it is essential to
understand the upcoming infected cases for several days.
That may help people and the government make pre-
decision before the pandemic to save live. In this paper,
we proposed a COVID-19 short-term forecasting model
using Linear Regression (LR), Least Absolute Shrinkage
and Selection Operation (LASSO) Regression, and
Support Vector Regression (SVR) to predict the next
seven days of COVID-19 infected cases in Bangladesh
during the pandemic situation. Here we considered data
from 8th May 2021 to 21st July 2021. We analyzed
different past data volumes for the model to understand
the impact of past data in the model. The result reveals
that Support Vector Regression (SVR) performance was
better than LR and LASSO in all aspects with high
accuracy. The performance also indicated that the high
volume of past data helps to increase prediction accuracy.
Keywords :
COVID-19, Short-term Forecast, New infected cases, Bangladesh, Supervised Machine Learning, LR, LASSO, SVR.
