An Explainable Machine Learning Framework for Breast Cancer Classification Using WBC and WDBC Datasets

Abstract

Breast cancer remains one of the leading causes of mortality among women worldwide, making early detection critically important for improving survival rates. This study presents an explainable machine learning (ML) framework for the classification of breast cancer using two widely recognized datasets: the Wisconsin Breast Cancer (WBC) and Wisconsin Diagnostic Breast Cancer (WDBC) datasets. The proposed approach integrates multiple ML algorithms, including K-Nearest Neighbors (KNN), Support Vector Machine (SVM), Random Forest (RF), XGBoost, and Artificial Neural Networks (ANN), to evaluate classification performance. The methodology involves data preprocessing, feature selection, model training, and evaluation using performance metrics such as accuracy and precision. In addition, Explainable Artificial Intelligence (XAI) techniques, including permutation importance, Partial Dependence Plots (PDP), and SHAP values, are applied to enhance the interpretability of the models and identify the most influential features. Experimental results indicate that KNN achieves the best performance on the WBC dataset with an accuracy of 97.7%, while ANN performs best on the WDBC dataset with an accuracy of up to 99%. The findings reveal that “bare nuclei” and “worst area” are the most significant features contributing to breast cancer classification. Furthermore, the study highlights a potential relationship between these features, offering deeper insight into the disease characteristics. The proposed framework demonstrates the effectiveness of combining machine learning with explainability techniques to support accurate and transparent medical diagnosis.