Utilization of Machine Learning Techniques for Prediction of COVID-19 Epidemic

ABSTRACT

COVID-19 pandemic is a deadly impact on the health and well-being of the world population. A developing country like Bangladesh has limited medical resources, and sometimes many people cannot get proper treatment in time. A continued increasing number of people tested positive for COVID-19 has caused a lot of strain on the governing bodies across the country, and they face difficulties to handle this situation. The aim of this work is to analyze the symptoms and predict the chances to get infected with COVID-19 disease. Five different machine learning algorithms are utilized to predict COVID-19 based on symptoms. Random forest, support vector machine, logistic regression, Gaussian Naive Bayes, and K-nearest neighbor algorithms have been used. We compare the performance before and after applying principal component analysis. The performance of K-nearest neighbor found the more accurate result before and after applying principal component analysis. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.