Abstract:
COVID-19 infections have become prevalent, prompting worldwide efforts to control
and treat the virus. Unfortunately, even after the invention of the vaccine so far, no
vaccine invention organization has claimed that their vaccine can completely prevent
Coronavirus and therefore the virus isn't going to be completely prevented. Since this
life-threatening virus has no specific and special treatment and it spreads very easily
and very quickly in human habitation. So, in an overpopulated and developing country
like Bangladesh in south Asia, it's very difficult to identify every infected person and
give them proper treatment for government and health workers. In recent years,
artificial intelligence and machine learning have achieved appeal as a part of enhancing
healthcare and research in general, especially in the field of the medical sector. To
predict "COVID-19", a wide range of machine learning approaches, applications, and
algorithms are developed. A machine-learning model is developed through which a
potentially infected individual can know how susceptible he/she is to become infected
with COVID-19 and their conditions. It may be very helpful for people to detect their
problem and get primary treatment from home until they reach the stage of going to the
hospital. This may make it possible to reduce the burden of health workers and the
government. To acquire the best potential result in this system, more advanced and
dynamic algorithms are required, such as K-nearest Neighbor, Decision Tree, Random
Forest, AdaBoost, XGBoost, Stochastic Gradient Descent, Linear SVC, Perceptron,
Naive Bayes, Support Vector Machines, Logistic Regression, Discriminant Analysis,
and other.
