A Predictive Model for Novel Corona Virus Detection with Support Vector Machine

Abstract

In this research, we have proposed machine-learning model for predicting coronavirus by employing support vector machine. The proposed model assists medical experts in the early detection of novel coronavirus disease (COVID-19) and helpful to make better decision during the coronavirus diagnosis. The objective of this study is to solve the problems caused by the wide-spreading coronavirus pandemic through machine learning approaches assisting the medical experts in detecting a novel coronavirus in the early stage. The detection of the coronavirus in early stage is vital in reducing the number of new cases of coronavirus through isolation of infected person from uninfected people. A support vector machine trained on the COVID-19 dataset collected from the online Kaggle data repository for coronavirus tested case dataset. Finally, we have analyzed the performance of the proposed model on coronavirus prediction with accuracy, precision recall, and receiver operating characteristic curve as performance metric in the evaluation of the model. The experimental test result reveals that the model has an accuracy score of 96.68% on detection of coronavirus.