The most prevalent oropharyngeal malignancy is tonsil cancer, and as cancers caused by the human papillomavirus (HPV) are becoming more common, their incidence is also significantly increasing recently. Several popular Machine Learning classifiers have been used to predict numerous types of cancer survivability. And it can make a substantial contribution to the field of cancer prediction. In accordance with that and finding the best effective models, we applied ten Machine Learning classifiers to the Surveillance, Epidemiology, and End Results (SEER) program database by choosing pertinent features to predict tonsil cancer survival status and evaluated the performance of each of these algorithms using accuracy, and F1 score, precision, recall, specificity, confusion matrix, and cross-validations. Additionally, we balance the Alive and Death classes using the Synthetic Minority Oversampling Technique (SMOTE). These methods have been used to analyze tonsil cancer patients’ chances of surviving and help doctors make a prognosis about the phase of the illness. According to our result, the Random Forest classifier outperformed others and obtained the highest score of 93.88%. Besides this, the Extra Tree classifier, and LGBM classifier also can be helpful in predicting the survivability of tonsil cancer patients. In addition, rather of referring to several publications, this resercah will also assist the researchers in quickly reviewing the relevant literature.