Elaboration of a Set of Independent Variables for Mathematical Models of Neural Networks in the COVID-19 Death Regression
COVID-19, Independent Variables, Neural Networks, Regression
We propose the elaboration of a set of independent variables, based on the analysis of phenomena, government interventions, and events that are important in the dissemination of the SARS-CoV-2 virus, whether positive or negative to the trend of the death curve. For this, isolated studies were carried out, based on machine learning methodology, of phenomena, government actions, and events to identify which ones should be selected and applied to neural network models such as LSTM. In the end, these isolated studies are compiled into a set of independent variables to regress the number of deaths caused by COVID-19 favoring higher accuracy and less standard deviation relatives to the real values. The main contributions are studies about the cause and relationship of phenomena, government actions, and events on the spread of the virus into urban conglomerates, such as cities and countries. The results of these studies can serve as auxiliary material to governs and governments in decision-making in the face of a pandemic situation such as the COVID-19 pandemic, or future situations of new pandemics.