COVID-19 pandemic has been going on for more than two years and an increasing number of deaths has been occurring. Ensemble learning techniques are effectively employed to predict the outcome of the patients with COVID-19. The mortality prediction of the COVID-19 patient is crucial to reduce the risk of imminent death as well as to apply effective clinical treatment strategy. In this study, we perform bagging and boosting methods to predict mortality of the patients with COVID-19. The six different decision tree methods, C4.5, Random tree, REPTree, Logistic Model Tree, Decision Stump, and Hoeffding Tree are employed for base learners in bagging and boosting. The results are obtained using a real-world dataset including information obtained from 1085 patients. Experimental results present that bagging using REPTree as a base learner achieves an accuracy of 97.24%. Furthermore, when we compare our results with other classification algorithms, the proposed method has a higher performance with respect to the accuracy, and presents an admirable performance.