Exploring Machine Learning Approaches for Breast Cancer Prediction: A Comparative Analysis with ANOVA-Based Feature Selection

Abstract

A widespread issue across the globe, breast cancer impacts women across diverse regions and populations. Early detection remains crucial for improving treatment outcomes and reducing mortality rates associated with the disease. Advancements in technology, especially in Machine Learning (ML), present promising opportunities to enhance the accuracy and effectiveness of breast cancer detection methods. The research carried out in this investigation involves a comparative analysis of three ML models (DT, ANN, SVM), utilizing the Wisconsin Diagnostic Breast Cancer (WDBC) dataset incorporating ANOVA for feature selection. The primary objective is to evaluate the effectiveness of these models in achieving precise and timely breast cancer detection. Through a comprehensive assessment, which includes common metrics, our findings underscore the superior performance of the SVM model, achieving a precision, recall, and F1-score of 98.59%. These results underscore the SVM's potential for accurate and early prediction of breast cancer using this dataset. This research contributes to advancing our understanding of machine learning methodologies in breast cancer diagnosis, emphasizing the significant role of technology and ML in facilitating early detection and improving patient outcomes.