Performance Comparison of Support Vector Machine (SVM) and Decision Tree C.45 for Breast Cancer Classification Model

Breast cancer remains a significant health challenge, with a high incidence and mortality rate globally, including Indonesia. This study aims to develop and compare the performance of machine learning models for breast cancer classification based on symptoms experienced by patients. The algorithms used in this research are Support Vector Machine (SVM) and Decision Tree C.45. Utilizing a dataset obtained from Kaggle, which was processed using the Synthetic Minority Oversampling Technique (SMOTE) to address class imbalance, the study employs Recursive Feature Elimination (RFE) for feature selection. The models were evaluated using accuracy, Fl score, and Matthews Correlation Coefficient (MCC) through k-fold cross-validation with k values of 3, 5, 7, and 10. Results indicate that the SVM algorithm consistently outperforms the Decision Tree C.45 across all metrics, achieving the highest accuracy of 96.64 % and demonstrating superior reliability and effectiveness in breast cancer classification. These results suggest that SVM is a promising early detection and treatment tool, potentially improving patient outcomes.