Comparative Study of Machine Learning Algorithms for Breast Cancer Diagnosis

Breast cancer remains a leading cause of cancer-related mortality among women globally. According to 2024 statistics, approximately 310,720 cases were reported in India alone, highlighting the urgent need for accurate and early diagnostic solutions. Machine Learning (ML) has become an effective approach in healthcare, enabling the development of predictive models for early detection and clinical decision support. This study presents a comparative analysis of several supervised ML algorithms, Logistic Regression, Support Vector Machine (SVM), k-Nearest Neighbors (KNN), Naive Bayes (NB), Random Forest (RF), and Extreme Gradient Boosting (XGBoost), using the Breast Cancer Wisconsin (Diagnostic) dataset. We applied preprocessing, stratified sampling, and hyperparameter tuning via GridSearchCV, followed by evaluation using accuracy, precision, recall, F1-score, and ROC-AUC. Results show that XGBoost achieved the highest test accuracy of 98.24%, outperforming SVM (97.5%), Random Forest and Logistic Regression (97.3%), KNN (93%), and Na¨ ıve Bayes (92%). These outcomes affirm the superior performance of ensemble learning in medical classification. This work demonstrates the promise of machine learning, particularly boosting techniques, in enhancing the reliability of breast cancer diagnosis.