Background: In the year 2020, coronavirus disease (COVID-19) has taken the proportion of a global pandemic. In this context, the effort of this work is to analyze, based on a data set of 180 countries, the dynamics and nature of global vulnerability to this pandemic.Methods: The results of Machine Learning, a resource in the field of artificial intelligence, were applied to two data sets obtained from the World Development Indicators (WDI) and the Worldometer website. The regression had its results generated by the algorithm C 4.5, Decision Tree. The first data set covered 47 variables on the economic activities of 180 countries. Of these, 45 were analyzed in a time series of 2017, 2018 and 2019 and were obtained from the WDI of the Economic Policy & Debt and Private Sector & Trade categories; 2 were collected from Worldometer reports on coronavirus, of the Total Cases categories and the total cases per 1 million population . The second data set covered 114 WDI variables from 180 countries, 49 of which were obtained from the Environment categories, 63 from the Health and Population fields, and 2 were the same as those from the first data set collected from Worldometer reports on coronavirus. Both data sets had the Total Cases per 1 million population variables as their main target attribute. The modeling of indicators related to economic activities, public health and population activities had as a basic assumption the fact that they express different forms and intensities of human interactions that remain as explanatory factors of the pandemic, since they interfere in the country's vulnerability to outbreaks of coronavirus . As a natural corollary of this basic assumption of research, we estimate a Vc number to measure the level of vulnerability of countries to COVID19.Findings: Applying the results of Decision Trees (C 4.5), we found six variables of greater relevance to form a typical profile capable of explaining the level of vulnerability of countries to coronavirus. The epidemic , as well as the geolocation of its epicenter, can be understood from the level of gross national income per capita, Industry as % of product domestic gross (GDP), International tourism (number of arrivals as a% of the total pop). Urban population (% of total pop.), Level of Diabetes prevalence (% of total pop on ages 20 to 79) and, Prevalence of HIV (% of total population ages 15-49). Aiming a simplified analysis, we estimate the levels of Vc vulnerability from the simple sum of these variables which equation corresponds, respectively, to Vc = Ic + Yc + Tc + Uc + Dc + Hc. According to our estimates, while in countries such as Denmark, Israel, Uzbekistan, Eq. Guinea, Gabon, Chile, Brazil, Paraguay, Czech Republic, Germany, Norway, and the USA the epidemic tends to be more severe, in countries like Zimbabwe, Gambia, Guinea, Ethiopia, Tanzania, Guinea-Bissau, Namibia, Burundi, and Rwanda it tends to advance at a slower pace.Conclusions: In summary, according to this study, the most advantageous measures to control the pandemic for society must consider the vulnerability of countries or regions based on their degree of development of the economy, industry, and tourism, as well as the presence of Diabetes and HIV. Due to the worsening of vulnerability due to pandemic epicenters proximity, our model also predicts a more severe increase in cases of COVID19 in African countries such as Gabon, Algeria, Senegal, Congo, Botswana, and Lesotho
