Feature Selection in Naïve Bayes for Predicting ICU Needs of COVID-19 Patients

COVID-19 is a global pandemic that requires coordinated response in all healthcare and national systems. Identifying patients at high risk of contracting the virus crucial to increasing awareness before become further infected by virus, which can cause severe respiratory illnesses requiring specialized care intensive units (ICUs). This study aims predict need for ICUs with virus. The predicted ICU requirements serve as reference hospitals meet needs patients. prediction performed using Naïve Bayes algorithm, particle swarm optimization (PSO) used obtain best accuracy values from Bayes. In initial testing, without feature selection resulted an rate 74.75%. Testing Bayes+PSO number PSO generations shows increases, also increases. 3000 population size 20 increase 80.95%. 40 1000 each 80.70%.