Title
Modified Machine Learning Techique for Curve Fitting on Regression Models for COVID-19 projections
Abstract
COrona VIrus Disease 2019 (COVID-19) is a disease caused by Severe Acute Respiratory Syndrome Corona Virus 2 (SARS-CoV-2) and was first diagnosed in China in December, 2019. Dr. Tedros Adhanom Ghebreyesus, World Health Organization (WHO) director-general on March 11th declared the COVID-19 pandemic. The cumulative cases of infected individuals and deaths due to COVID-19 develop a graph that could be interpreted by an exponential function. Mathematical models are therefore fundamental to understanding the evolution of the pandemic. Applying machine learning prediction methods in conjunction with cloud computing to such models will be beneficial in designing effective control strategies for the current or future spread of infectious diseases. Initially, we compare the trendlines of the following three models: linear, exponential and polynomial using R-squared, to determine which model best interprets the prevailing data sets of cumulative infectious cases and cumulative deaths due to COVID-19 disease. We propose the development of an improved mathematical forecasting framework based on machine learning and the cloud computing system with data from a real-time cloud data repository. Our goal is to predict the progress of the curve as accurately as possible in order to understand the spread of the virus from an early stage so that strategies and policies can be implemented.
Year
DOI
Venue
2020
10.1109/CAMAD50429.2020.9209264
2020 IEEE 25th International Workshop on Computer Aided Modeling and Design of Communication Links and Networks (CAMAD)
Keywords
DocType
ISSN
covid19,coronavirus,machine learning,cloud computing,regression,forecast,epidemic,pandemic,curve fitting
Conference
2378-4865
ISBN
Citations 
PageRank 
978-1-7281-6339-0
0
0.34
References 
Authors
0
6