Air Quality and COVID-19 Patient Conditions in Jakarta: A Comparative Analysis of Classification Algorithms

The COVID-19 pandemic has become a global challenge, with environmental factors such as air quality contributing to disease severity. This study analyzes the relationship between air pollution parameters (PM2.5, PM10, NO2, SO2, CO, and O3) and COVID-19 patient conditions in Jakarta, categorized into three groups: positive, recovered, and deceased. A comparative evaluation was conducted using five classification algorithms: Na¨ıve Bayes, Random Forest, k-Nearest Neighbors (kNN), Decision Tree, and Support Vector Machine (SVM). The results show that kNN achieved the highest accuracy of 80.71%, while Na¨ıve Bayes obtained the highest recall of 91.83% and a precision of 80.75%. This study contributes by evaluating the effectiveness of classification techniques in mapping the impact of air quality on patient conditions and by identifying the most accurate predictive model. The findings suggest that classification methods can serve as reliable predictive tools to assess the health impacts of air pollution on the population.