Machine Learning Models for Prediction of COVID-19 Infection in North Macedonia

The COVID-19 pandemic, caused by the
SARS-CoV-2 virus, has emerged as one of the most
significant global crises of this century, with severe
health and socio-economic impacts worldwide. Existing
research has highlighted the critical role of
comorbidities in influencing COVID-19 outcomes, but
effective prediction models remain a challenge. This
study investigates the potential of machine learning
algorithms to predict the outcomes of COVID-19 based
on patients' comorbidities. The algorithms K-Nearest
Neighbors, Decision Tree, Logistic Regression, and
Random Forest are applied to an epidemiological
dataset comprising only positive COVID-19 cases,
obtained from the Public Health Institute of North
Macedonia. Additionally, two ensemble learning
techniques, XGBoost and RUSBoost, are used to
enhance prediction accuracy. The models achieved high
accuracy of 90% across the various algorithms. These
findings suggest that machine learning models can be an
effective tool for predicting COVID-19 outcomes,
especially when comorbidity data is available.