Use of Deep Learning to Identify COVID-19 through DNA/RNA Sequencing

COVID-19, or coronavirus, is a virus that causes a pandemic where the severe acute respiratory syndrome virus 2 (SARS-Cov-2) affects the respiratory organs such as the lungs. As of 4th February 2024, there are over 774 million confirmed cases of COVID-19 with more than 7 million deaths have been reported globally. Although the pandemic has slowly become controlled over the years, there is room to find a more effective way to cure COVID-19. The aim of this study is to find out an effective machine learning model to identify COVID-19. The dataset used for this study is the GSE152418 from GEO. After testing 4 proposed models, this study shows that RFE-SVM is the most effective method for identifying COVID-19, with an accuracy of 0.8733, precision of 0.9021 and a recall score of 0.8571. Due to the high score across all metrics, the RFE-SVM model is able to be used and improved for future studies.