Real-time Estimation of Global CFR Ascribed to COVID-19 Confirmed Cases Applying Machine Learning Technique

The COVID-19 is declared as a public health emergency of global concern by World Health Organisation (WHO) affecting a total of 201 countries across the globe during the period December 2019 to January 2021. As of January 25, 2021, it has caused a pandemic outbreak with more than 99 million confirmed cases and more than 2 million deaths worldwide. The crisp of this paper is to estimate the global risk in terms of CFR of the COVID-19 pandemic for seventy deeply affected countries. An optimal regression tree algorithm under machine learning technique is applied which identified four significant features like diabetes prevalence, total number of deaths in thousands, total number of confirmed cases in thousands, and hospital beds per 1000 out of fifteen input features. This real-time estimation will provide deep insights into the early detection of CFR for the countries under study. CFR[Formula]as suggested by (Boldog et al., 2020, Chakraborty et al. 2019, Russell et al., 2020) Diabetes Prevalenceproportion of a population who have diabetes in a given period of time. Stringency Indexit provides a computable parameter to evaluate the effectiveness of the nationwide lock down in a particular country. GDP Per Capitait is a metric that breaks down a countrys economic output per person and is calculated by [Formula] Population Densityit is a measurement of population per unit area. It refers to the number of people living in an area per square kilometre. HDIit is a statistic composite index of life expectancy, education (literacy rate, gross enrolment ratio at different levels and net attendance ratio) and per capita income indicators which are used to rank countries into four tiers of human development.