Real-time Breast Cancer Prediction in Industrial Healthcare: A Machine Learning Operationalization Approach

Abstract

Timely identification of breast cancer greatly enhances patient outcomes. This research assesses the effectiveness of five traditional machine-learning classifiers Logistic regression, decision tree, random forest, and K-Nearest Neighbors (KNN), Support Vector Machine (SVM) for binary classification of tumors (benign versus malignant) utilizing the Wisconsin Breast Cancer Wisconsin dataset. Models are checked through accuracy, confusion matrices, ROC curves, and classification reports. Our tests indicate that SVM and logistic regression reach the greatest accuracy (approximately 98.2% and 97.4% respectively) with elevated ROC AUC, in agreement with prior research.