TY - JOUR AU - Darwin Patiño Pérez AU - Celia Munive Mora AU - Lorenzo Cevallos-Torres AU - Miguel Botto-Tobar PY - 2021/09/30 Y2 - 2024/03/29 TI - Predicting the Effectiveness of Rapid Tests Performed to Patients with COVID-19 through Linear Regression and Random Forest JF - Ecuadorian Science Journal JA - Ecuad. Sci. J VL - 5 IS - 2 SE - Research Paper DO - 10.46480/esj.5.2.108 UR - https://journals.gdeon.org/index.php/esj/article/view/108 AB - The rapid spread of SARS-CoV2 (COVID-19) has caused a collapse of health systems worldwide, so a strategy to control the spread is the timely detection of the virus through rapid tests, which allows acting and thus giving a timely treatment that reduces its spread. With the help of artificial intelligence techniques, within the subfield of machine learning or machine learning, there have been significant advances that allow speeding up the analysis of large volumes of data. This study aims to determine the effectiveness of rapid tests in detecting covid-19, using machine learning, applying a methodology that involves the creation of linear regression and Random Forest models with the Python programming language. In the methodology used, the models were created, which were then defined and trained, and after performing the tests and predictions, the validation metrics determined the precision and effectiveness of these models. From the results obtained, it is concluded that the random forest model is good since it provided a precision of 61%, but with the linear regression model, it was determined that it has a precision level of approximately 90%, so finally, with these results, health professionals will be able to make reliable predictions regarding the effectiveness of rapid tests as a mechanism that will help to quickly detect the presence of the virus and thus reduce the spread of the virus. ER -