PREDICTING THE GROWTH AND TREND OF THE COVID-19 VIRUS USING MACHINE LEARNING

Bharti Sahua, Dr. Bhagwan Phulpagar Digambar Jadhav

Abstract

The global SARS-CoV-2 outbreak caused by the COVID-19 Coronavirus has been catastrophic. COVID19's cumulative incidence is rising at an alarming rate. Tracking the disease, predicting its progress, and
developing strategies and regulations to control the epidemic are all tasks that can greatly benefit from the
use of Machine Learning (ML) and Cloud Computing. This research makes use of a sophisticated
mathematical model to examine and predict the epidemic's progress. An improved model based on ML
has been used to estimate the risk of COVID-19 in different nations. We show that iterative weighting for
fitting may lead to a better fit when building a prediction framework using the Generalized Inverse
Weibull distribution. This has been implemented on a cloud computing platform for improved and more
timely prediction of the epidemic's development propensity. Proactive responses from both the
government and the people can greatly benefit from a data-driven strategy with the level of precision
shown here. Finally, we suggest a number of avenues for future study and actual implementation. From
this review of the relevant literature, we were able to choose a group of prediction-friendly algorithms,
including SVMs, RFs, and ANNs. The selected algorithms' performances are compared in order to
determine which method provides the most accuracy. To determine the significance of each feature in the
context of the forecast, importance values are calculated. The use of Machine Learning for COVID-19
prediction has the potential to increase the rate at which diseases are diagnosed, which in turn would
reduce mortality rates. Based on the experimental data, it was determined that the artificial neural network
outperformed the other algorithms.