Machine Learning and Bayes Probability For Detecting Camouflaged Mini Pandemic at the Waves of Covid-19

Abstract

This paper present a methodology based at Machine Learning and a theory backed by the Bayes probability to identify rare strains that might not be in coherence with the corona virus. By using the criteria of Tom Mitchell applied on the data belonging to 2021–2022 period, the distributions of infections registered at the beginning of 2022 would not be in accordance to waves of pandemic as seen at 2020 and 2021. Therefore, algorithm of Machine Learning has yielded that the so-called Omicron variant would no be coherent with known mutations neither exhibiting same pattern of previous waves of pandemic. This creates a space to speculate about the origin of new strains that are camouflaged to central corona virus. From the results of this work, it is observed that Omicron might have nothing to do with Covid-19 pandemic, instead it have triggered a small pandemic of short duration as validated by global data.