Prediction of Mortality Rate among Various Age Groups of Covid-19 Patients in India Using Naive Bayes Algorithm for Effective Decision Making

Abstract

The noble corona virus or COVID-19
has spread in almost 213countries across the globe and
its wrath has affected more than 7 million people and
has claimed about 429067 lives by june 1st week. Our
country, India stands at the 4th position among the
affected countries worldwide. In India only there are
more than 3 lakhs positive cases of COVID-19 out of
which 8895 people lost their precious lives to this deadly
disease. In a densely populated country like India, where
about 464 people live in 1 km2 of area and about 550
hospital beds are available per 1 million populations, we
will soon be run out of beds if the confirmed cases
increase at the present rate. So, to save the lives of
people the government and hospital personnel must take
some decisive measures to avoid the crowd and provide
proper healthcare facilities to the infected people. Here
we have tried to help the health care worker and hospital
staff to make decisions to provide emergency medical
facilities to people on a priority basis when the system is
in overcrowded condition. It will help them to make
quick and right decision to hospitalise the more
vulnerable patients with high mortality rates, such as
older age group and patients with pre-existing medical
conditions. In this model, we have used data set of
27891 confirmed cases of covid-19 people and
developed a predictive model using various machine
learning algorithms, such as J48 classifier, Naive Bayes,
ranker search methods, and Random forestto predict the
death rate among the various age group of infected
people. The outcome gives around 94% accuracy in
predicting the mortality rate.