Performance Analysis of Machine Learning Algorithms in Detection of COVID-19 from Common Symptoms

Abstract

COVID-19 disease caused by the SARS-CoV-2 virus has spread rapidly around the world and turned into a global epidemic. Machine learning methods, a powerful tool against COVID-19, are effectively used to combat this epidemic. In this study, six machine learning algorithms, which are Artificial Neural Network, Naive Bayes, Support Vector Machine, k-Nearest Neighbor, Random Forest and AdaBoost methods, are compared and evaluated for classifying COVID-19 cases on the data including the main symptoms of 10.000 people, which are fever, cough, sore throat, shortness of breath, headache as well as gender, age and contact data with the infected person. Experimental results conduct that artificial neural network achieves the best accuracy of %87. On the other hand, the other machine learning methods reach an accuracy of 84% on average for detecting COVID-19 cases from common symptoms.