The Basic Study on Support Vector Machine (SV...

The Basic Study on Support Vector Machine (SVM) on Covid 19 Worldwide Dataset

Author Name : Rukhsun Ara Parvin, Tiyasha Dhara, Arpan Adhikary

ABSTRACT

We have proposed the work using Support Vector Machine to predict the number of death, number of recovery, the number of daily cases, number of active cases and the total number of cases by COVID. The data we have used is collected in between the time of 22^nd January, 2020 and 31^th January, 2021. By the time, total number of confirmed cases was 109,217,366 with 2,413,912 total number of deaths and 138,688,383 of recovered cases. This model has been developed in python of version 3.8.1 to get the predicted values using Support Vector regression model with Radial Basis Function as the Kernel and 10% confidence interval for curve fitting. The data has been split into train and test set with test size of 40% and training size of 60%.  The model performance parameters are calculated as mean square method, root mean square error, regression score and percentage accuracy. The model has 65% accuracy (as we have used only 2 iteration) in predicting deaths, recovered, cumulative number of confirmed cases and 87% accuracy in predicting daily new cases. The results suggest a Gaussian decrease of the number of cases and could take another 3 to 4 months to come down the minimum level with no new cases being reported. The method is very efficient and has higher accuracy than linear and polynomial regression.