Regresar

Data Mining Application. Case Study: Covid-19 in Ecuador

Abstract:

The health emergency due to the COVID-19 pandemic and has been necessary to be treated with immediate and integrated response to all organizations involved in order to achieve data sharing on this and future global pandemics of rapid spread. This study focuses on predicting the incidence of COVID-19 in Ecuador by data mining the records provided by public institutions of the Ecuadorian state with official information on COVID-19 in Ecuador. We experimented with regression and long-term memory models, obtaining as a result the optimal model to estimate the number of positive cases of COVID-19. For the models used, the quadratic error was used as a performance metric. From the analysis of the data on COVID-19 in Ecuador, the polynomial regression model predicted the incidence with a quadratic error of 0.86, the most effective factors being the incidence of previous days and the number of population in each of the affected provinces.