Developing a Model to Predict the Effectiveness of Vaccination on Mortality Caused by COVID-19

The Coronavirus Disease 2019 (COVID-19) pandemic highlighted the urgent need for effective vaccination strategies to control the virus’s spread and reduce mortality. Machine learning (ML) algorithms offer promising tools for predicting vaccine effectiveness and aiding public health decisions. This study explores the application of various ML techniques, including artificial neural network (ANN), decision tree (DT), K-nearest neighbor (KNN), random forest (RF), and support vector machine (SVM) to model and forecast the impact of vaccination on COVID-19 mortality. The algorithms were evaluated using accuracy, precision, recall, specificity, F-measure, and area under the curve (AUC) metrics. The findings revealed that DT outperformed other ML algorithms, achieving the highest metrics across multiple evaluation criteria. It recorded an accuracy of 92.27%, precision of 92.54%, recall of 91.95%, specificity of 87.92%, F-measure of 92.24%, and an AUC of 94.50%, highlighting its exceptional predictive performance. Moreover, DT demonstrated this high level of accuracy while maintaining minimal computational time. These findings suggest that ML models, particularly DTs, can be valuable in assessing vaccine effectiveness and informing health strategies against COVID-19.