Machine Learning Classifier Model for Prediction of COVID-19

Abstract

COVID-19 pandemic has become a major threat to the world. In this study a model is designed which can predict the likelihood of COVID-19 patients with maximum accuracy. Therefore three machine learning classification algorithms namely Decision Tree, Naive Bayes and Logistic Regression classifier are used in this experiment to detect Covid-19 disease at an early stage. The models are trained with 75% of the samples and tested with 25% of data. Since the dataset is imbalanced, the performances of all the three algorithms are evaluated on various measures like F-Measure, Accuracy and Matthews Correlation Coefficient. Accuracy is measured over correctly and incorrectly classified instances. All the analyses were performed with the use of Python, version 3.8.2. Receiver Operating Characteristic (ROC) curves are used to verify the result in a proper and systematic manner. This framework can be used, among other considerations, to prioritize testing for COVID-19 when testing resources are limited.