<p>COVID-19 was a severe health disaster that caused substantial health and socioeconomic disruptions with high mortality rates worldwide. Accurate diagnosis of COVID-19 is crucial for preventive and effective clinical decisions, although a delay in diagnosis may result in high mortality rates. This necessitates a need to improve pandemic response strategies and optimize clinical outcomes in order to minimize the mortality due to COVID-19. Machine learning models can be used in early mortality prediction due to COVID-19. This paper attempts to design a classification-based machine learning model for predicting mortality due to COVID-19. Several classification techniques such as K-NN, Naïve-Bayes, Decision tree, Random Forest, XGBoost and CatBoost were used to design COVID-19 mortality prediction models using the COVID-19 patients pre-condition dataset published by Mexican government. Performance of these models were evaluated using metrics such as Accuracy, Precision, Recall, F1-score, MAE, RMSE and AUC-ROC SCORE. The results showed that XGBoost-based COVID-19 mortality prediction model performed the best. The use of this model would offer valuable insights and would result in effective and early prediction of mortality in COVID-19 patients.</p>