The challenge of distinguishing genetic mutations that contribute to tumor growth is crucial in cancer treatment. Cancer is responsible for millions of deaths annually, hence the need for early detection of tumors to improve treatment efficacy and survival rates. However, manual classification is prone to errors and inefficiencies due to human limitations and the complexity of domain knowledge, leading to time-intensive processes. In response, machine learning models improve accuracy and efficiency for cancer prognosis and prediction. However, the lack of theoretical understanding of algorithms may limit the interpretability and applicability of results, where insights into model prediction are crucial to making informed decisions, especially in the biomedical domain. To address these challenges, our study employed four supervised machine learning algorithms, namely Support Vector Machine (SVM), Naïve Bayes (NB), Logistic Regression (LR), and Random Forest (RF). The performance of these algorithms was assessed using log-loss and misclassification rates. Logistic regression emerged as the optimal classifier with a log loss of 1.0125 and a misclassification rate of 30.97%.