BREAST CANCER CLASSIFICATION USING MACHINE LEARNING : A DEPLOYABLE AND INTERPRETABLE DIAGNOSTIC SUPPORT SYSTEM

Abstract

Breast cancer ranks among the most common cancers affecting women globally. The timely identification of breast cancer is vital for enhancing patient survival rates. This research offers a comparative evaluation of three machine learning techniques—Logistic Regression, Decision Tree, and Support Vector Machine (SVM)—to differentiate between malignant and benign breast cancer using the Breast Cancer Dataset. Following data preprocessing, correlation-based feature selection, and model assessment, Logistic Regression demonstrated the highest accuracy at 98.25%, outpacing the Decision Tree and SVM models. The findings highlight the effectiveness of logistic regression as a dependable predictive tool for breast cancer diagnosis.