TREND ANALYSIS AND COMPARISON OF COVID - 19 CASES BEFORE AND AFTER ADMINISTRATION OF VACCINE USING MACHINE LEARNING

ABSTRACT

In this study we draw a comparison between the trends in the growing cases of novel coronavirus after the administration of vaccine doses. We compare two scenarios where how the trends have changed after the vaccine has been administered and how the trend would have looked if there were no vaccines present. This study can be used to determine the early changes that the vaccines have brought about in the trends and how much reliability do they show in preventing the cases from rising further. The predictions are made using a Weibull based Long-Short-Term-Memory approach which is also being used by the National Health Service of the UK on a dataset that takes into account features like age groups, air traffic, developmental index of the country, average temperatures of a country, which are detrimental in determining the rate of infection and deaths accurately. The model is tested on data gathered from multiple countries and the results are drawn after analyzing the result for each country as an individual entity for the conclusion to be reliable. With an increasing market competition and not so long testing period given to these vaccines which have made it to the common masses we feel this study can help predict how effectively the vaccines will be able to improve immunity against this virus and is it a viable option to invest such large capital in development and purchase of these vaccines preferring it over the organically decreasing curve following the traditional methods and natural processes. © 2021 IEEE.