Attribute selection based on genetic and classification algorithms in the prediction of hospitalization need of COVID-19 patients

Abstract

The COVID-19 pandemic has been pressuring the whole society and overloading hospital systems. Machine learning models designed to predict hospitalizations, for example, can contribute to better targeting hospital resources. However, as the excess of information, often irrelevant or redundant, can impair the performance of predictive models, we propose in this work a hybrid approach to attribute selection. This method aims to find an optimal attribute subset through a genetic algorithm, which considers the results of a classification model in its evaluation function to improve the hospitalization need prediction of COVID-19 patients. We evaluated this approach in a database of more than 200 thousand COVID-19 patients from the State Health Secretariat of Rio Grande do Sul. We provided an increase of 18% in the classification precision for patients with hospitalization necessities. In a real-time application, this would also mean greater precision in targeting resources, as well as, consequently and mainly, improved service to the infected population.