Analysis of Covid-19 using machine learning techniques

ABSTRACT

Covid-19 is caused by a newly detected coronavirus (SARS-CoV-2). It is a respiratory infection that usually spreads from individual to individual through sneezing or coughing. The disease, which was first detected in the province of Wuhan, China, had effected more than one continent and was declared as a pandemic by the World Health Organization (WHO). The pandemic has affected health, social, economic, and psychological segments of life for billions of people. Though vaccines have been developed and are made available, we are still prone to the virus, which is similar to any other flu. This chapter presents an analysis of the symptoms of the disease and identifies significant symptoms that impact the cause of the illness. Machine learning techniques like multiple regression, support vector machine (SVM), Decision Tree, Random Forest, and Logistic Regression are applied to understand the evaluation with respect to the measures like coefficient of determination, and mean-squared error. Hypothesis testing is used to determine whether at least one of the features is useful in the diagnosis of the disease. Further feature selection process is used to identify the most significant symptoms that will cause the virus. Different visualization methods are used to figure the substantial reasoning from the model's prediction and perform analysis on the results obtained. © 2023 Elsevier Inc. All rights reserved.