Since the start of 2020, the novel coronavirus has spread extensively, prompting numerous efforts to develop vaccines for patient recovery. It is evident now that a swift solution is required worldwide to control the spread of COVID-19. Amidst the COVID-19 pandemic, vaccines have been developed and granted emergency authorization to mitigate the occurrence of COVID-19. Extensive evidence supports the high efficacy of these vaccines in preventing severe illness, hospitalization, and fatality resulting from COVID-19. They serve as a vital resource in curbing the virus's transmission and ultimately ending the pandemic. Non-clinical approaches, such as data mining and machine learning techniques, hold promise in alleviating the strain on healthcare systems and providing an optimal diagnosis of the pandemic. The dataset contains 36 states/UTS, total cases, death cases, and total dose. In this paper, Machine Learning (ML) algorithms were employed to analyze and predict the COVID19-related dataset for finding the suitable parameters for prediction using M5P and Random Forest. The performance and their accuracy of these algorithms was evaluated using precision, recall, accuracy, and f-measure metrics. Numerical illustrations were also provided to prove the proposed research.