COVID-19 pandemic disease spread by SARS-COV-2 single-strand structure RNA virus belongs to the 7th generation of the coronavirus family. Following an unusual replication mechanism, its extreme ease of transmissibility has put many counties under lockdown. With a cure for the infection uncertain in the near future, the pressure currently lies in the current healthcare infrastructure, policies, government activities, and behaviour of the people to contain the virus. This research seeks to understand the spreading patterns of the COVID-19 virus through exponential growth modelling and identifies countries that have showed an initial sign of containment until 26th March 2020. Post identification of countries that have shown an initial sign of containment, predictive supervised machine learning models were built with infrastructure, environment, policies, and infection related independent variables. For the purpose, COVID-19 infection data across 42 countries were used. Logistic regression results shows a positive significant relationship of healthcare infrastructure and lockdown policies on the sign of early containment in countries. Machine learning models based on logistic regression, decision tree, random forest, and support vector machines were developed and are seen to have accuracies between 76.2% to 92.9% to predict early sign of infection containment. Other policies and activities taken by countries to contain the infection are also discussed.