Application of Machine Learning for the Prediction of Covid19 Through Classification Techniques and Supervised Learning

Authors

Molina-Calderon, Miguel

Crespo-Mendoza, Roberto

Reyes-Cun, Ericka

Patiño-Perez, Darwin

Burgos-Robalino, Freddy

Sarmiento-Barreiro, Liliana

Botto-Tobar, Miguel

Abstract

In January 2020, in the city of Wuhan in China, the appearance of a highly dangerous disease for humans, cataloged as COVID-19 and caused by a virus called SARS-CoV-2, was announced; From there, it spreads rapidly from Asia to Europe and then spreads throughout the American continent, causing a pandemic that to date has caused irreparable damage. The symptoms of the disease are initially similar to those of the flu, but are accompanied by other characteristics such as sweating, dry throat and loss of taste. Currently, a series of variants are registered, each of which, due to its symptoms, have generated treatments, but if they are not identified in time, they continue to cause death in all parts of the world. With the historical data registered in the portals endorsed by the world health organization and through the application of artificial intelligence techniques, a prediction model has been created, which using supervised learning has learned to identify the presence of the disease, as well as to predict its spread. In supervised machine learning, there are classification models for prediction, including Decision Trees and Gradient Boosting, which according to many studies reflect versatility, speed, reliability and efficiency to perform predictive analysis that would help health specialists provide fast and reliable treatment to all those affected by decongesting the care areas.