![Sample our Mathematics & Statistics journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5943/TOC-Subject-Sample-2021-Mathematics-Statistics.jpg"})
Abstract
COVID19 is a contagious ailment, which is emanated from the city of Wuhan (Hebei district), China in late of 2019, genesis from newly come across corona virus. Humans are contaminated with the COVID-19 virus and experiencing the clement to passable respiratory ailment, recuperated without necessary special treatment. In this study, a systematic review of prediction of COVID-19 using machine learning and big data is accomplished, by considering all the related imperative features, excluding CT scan and X-Ray Images data sets, with all the available related articles around the globe. The summary concluded that the prediction patterns for some algorithms are not satisfactory, the predicted value and actual values are inversed, and meanwhile some algorithms resulted with good accuracy and lower errors. No study is done to target Indian data set including population index, as India is a densely populated country holding 17.7% of world’s population. With respect to this, our study included the implementation of two classification algorithms for the Indian data of COVID-19 cases from 30th January 2020 to 30th May 2020 including population index state wise. Furthermore, the results are commensurately equal with both the implemented algorithms. Bayes point machine algorithm and logistic regression algorithm are cross validated with 10 cross folds over the data and the maximum accuracy achieved is 99.6% and 99.4% respectively. At the same time, 7th cross fold of Bayes point machine manifested the worst accuracy with lowest precision, recall and f-score while logistic regression seems to be clement with all the 10 cross folds. Prediction of future diegesis for COVID-19 can serve the medical decision making especially when it is evoking attention immediately.
Subject Classification: