Abstract:
COVID-19 disease has almost been witnessed by all countries of the world in 21st Century and is now declared as pandemic by World Health Organization (WHO). It is still a growing challenge due to testing of vaccines to cure this disease and it can be predicted that this process may still take quite a lot of time in providing to masses. In Pakistan, first case of Corona virus was reported on 26 February 2020, afterwards COVID-19 infections spread at quite a high pace in the year 2020. In 2021, COVID-19 is still a growing challenge for Pakistan and strict measures needs to be taken by health sector of Pakistan to abate the risk of communication. Machine learning techniques are used to help describe the size of an outbreak and the rate of spread of an infection in a population. Some Machine learning techniques were employed, around the world, to estimate the severity of the COVID-19 pandemic. Trend of this disease has shown varying behavior in almost all affected countries, therefore 1 method suitable for a specific country will not likely work for another country. It has been proved from deduced graphs that different degrees of polynomial regression model are working for different countries COVID-19 data for the same time period i.e. 13 months.Moreover, COVID-19 curve of a country has massive fluctuations, which depicts various feature sets are playing their role, leading to increase and decrease of cases. Therefore, there is a need to evaluate Government policies, Environmental and cultural feature sets playing their role in rise and fall of severity of disease. In Pakistan, it’s been 1 year still the country is not out of pandemic dangers, so we feel it is a good time to study the trend of feature sets which have affected the trend criticality to reflect on the predictions of pandemic spread. In order to build SOPs and future measures there is a need to observe and analyze the Machine Learning (ML) techniques for COVID-19 trend prediction and compare them on basis of performance parameters. This work will not only help the country's on-going efforts towards controlling the coronavirus, but it has also proposed a framework which will work as the basis of any prediction systems for such disruptive events in our society.
