ABSTRACT
There have been significant revolutions in various fields like medical and education on account of improved technological advancements. Furthermore, there have been numerous cases where Machine Learning has been of great help to healthcare by analyzing data and in decision making. Early diagnosis of Covid will help reduce the transmission rate and prevent an outbreak or slow down its spread. COVID-19 is a pandemic which is spreading really fast, affecting and killing millions around the globe and this needs to be addressed soon. Big data has been growing rapidly and there are many public datasets available related to COVID-19. ML could aid in the detection of the disease to bring the current chaotic situation under control. Various machine learning algorithms have been applied in this paper to build the most accurate model that can analyses symptoms of a person and predict if they are covid positive or not using a dataset from Kaggle. The performance of each model was analyzed according to different scoring metrics like accuracy measures, R squared, Precision, ROC curve and on how long the model took to be trained. It can be inferred from this paper that Decision Tree Classifier surpasses all the other algorithms by 98.29% accuracy. © 2022 IEEE.