Prediction of COVID-19 By Applying Supervised Machine Learning Techniques


  • Usman Shabbir
  • Mustafa Shakir



Bagging Tree, Intensive Care Unit, Machine Learning, Support Vector Machine, Random Forest


In the era of modern lifestyle, humans going to modernize everything with innovative brains finding the solution for the latest pandemic of this century which is COVID-19. The diagnosis of COVID-19 is a critical and slow process, to tackle this we decided to diagnose COVID-19 quickly by using machine learning (ML) due to its prediction for time-saving. Therefore, in this research, a patient admitted to the intensive care unit (ICU) has vital signs such as heart rate, oxygen saturation (SPO2), blood pressure, and respiratory rate using wearable sensors which are the main vital signs of COVID-19. the predictive models using machine learning algorithms support vector machine (SVM), random forest (RF), and bagging tree (BG)  for more accurate predictions. The proposed prediction model results used dataset, and models trained with Support Vector Machine in AUC=  0.9.1, ROC = 0.90 curve, Random Forest AUC = 0.920, and ROC = 0.91, and Bagging Tree AUC = 0.913, and ROC = 0.91 which shows the prediction accuracy. This research indicates machine learning's capability to predict COVID19 quickly instead of conventional methods that have previously been adopted in pandemics such as influenza.