Non-invasive prediction mechanism for COVID-19 disease using machine learning algorithms

Publication Name

International Journal of Critical Infrastructures

Abstract

This paper has focused on developing a model to detect non-diagnostically whether the person is infected with the COVID-19 disease using all relevant symptoms and details mentioned by the person and then comparing it with a pre-defined dataset of positive cases using machine learning. Different models have been developed to predict the same but none of them focused on the detection of COVID-19 based on symptoms. In a developing nation with a huge population, where the diagnostic availability is scarce, just scanning the body temperature will not help in detection of COVID-19 of a particular individual. This paper presents a model that can predict COVID-19 cases without any testing kit to an accuracy of 99.30%, performing better than other similar approaches with objective to put forward a method that can reduce the need of producing testing kits and also the need to wait for hours before we get the results.

Open Access Status

This publication is not available as open access

Volume

20

Issue

2

First Page

111

Last Page

124

Share

COinS
 

Link to publisher version (DOI)

http://dx.doi.org/10.1504/IJCIS.2024.137406