Comparative Analysis of Random Forest, Support Vector Machine, and Logistic Regression Algorithms in Breast Cancer Classification Using the Wisconsin Breast Cancer Dataset

Abstract

Breast cancer is one of the leading causes of cancer death in women worldwide, so early detection is an important factor in improving patient survival. This study aims to compare the performance of Random Forest, Support Vector Machine (SVM), and Logistic Regression algorithms in breast cancer classification using the Wisconsin Breast Cancer Dataset (WBCD). The dataset used consisted of 569 patient medical records with 30 numerical features obtained from fine needle aspiration (FNA) examinations. The research methodology includes data collection, data cleaning, label encoding, preprocessing using StandardScaler, and dividing the dataset into 80% training data and 20% test data. Model performance evaluation was carried out using accuracy, precision, recall, F1-score, confusion matrix, and Area Under the Curve (AUC). The results showed that Random Forest and SVM obtained the highest accuracy of 97.37%, while Logistic Regression achieved an accuracy of 96.49%. Random Forest and SVM produce a 100% precision score for the ferocious class, which means there are no false positive predictions. All three algorithms achieved the same recall value of 92.86% for malignant cases, reflecting its good ability to detect breast cancer. In addition, all models obtained an AUC value above 0.99, indicating excellent classification performance. Overall, Random Forest and SVM show higher performance consistency than Logistic Regression, so both can be considered as effective approaches in supporting early detection of breast cancer systems.