ABSTRACT
The Coronavirus disease (COVID-19), which emerged in 2019 and affected the world, has caused millions of people to be infected with the virus and hundreds of thousands of deaths. This disease has brought heavy workload to doctors and healthcare workers. This workload will be mitigated by machine learning and the development of computer-aided diagnostic systems. Any scientific study about the disease will help get rid of this disease as soon as possible. This chapter proposes a combination of feature extraction by using histogram of gradients and principal component analysis. Two data sets were used in this study: COVID-19 (−) and COVID-19 (+). K-nearest neighbor (KNN) and support vector machines (SVM) were applied as classifiers. It was found that the SVM classifier had a much better performance with an accuracy of 89.63% than the KNN classifier that gave an accuracy of 81.31%.
