COVID-19 Prediction Applying Machine Learning and Ontological Language

COVID-19, an infectious illness which has evolved into multiple varieties, has become a global epidemic that requires prompt diagnosis. In order to assist practitioners in predicting the existence of COVID-19, researchers are evaluating enormous amounts of complex medical data by combining different statistical and machine learning methodologies. In this study, we proposed a new method for anticipating Coronavirus illness, that combines ontology and machine learning to produce a powerful ontology-based model that can accurately foretell whether or not COVID-19 is present according to the symptoms, and thus provide an early identification. The process involves compiling the decision tree algorithm’s rules for differentiating virus-infected persons and then utilizing SWRL to apply these rules to the ontology reasoner. The ontology model achieves a high classification performance of 96% when compared to the decision tree model.