<![CDATA[Breast cancer is one of the leading causes of death among women worldwide. A major challenge in modeling patient survivability prediction is imbalanced data, where the number of surviving patients significantly outweighs the deceased ones. This study aims to compare the performance of three machine learning algorithms: Logistic Regression, Support Vector Classifier (SVC), and Gradient Boosting Classifier, in predicting patient survivability status. To address the class imbalance issue, Random Over Sampling (ROS) technique was applied during the data preprocessing stage. The methodology includes categorical data encoding, resampling, and model evaluation using accuracy, precision, recall, and F1-score metrics. Experimental results show that the application of ROS successfully balanced the class distribution. Among the three models tested, the Gradient Boosting algorithm demonstrated the best performance compared to linear and vector-based models. This study provides insights into the importance of handling imbalanced data to improve the accuracy of AI-based medical diagnoses.]]>
Copyrights © 2026
