<![CDATA[Breast cancer is one of the diseases with a high mortality rate in women, so early detection is crucial to increase the chances of recovery. Unfortunately, conventional methods of diagnosis still rely on the interpretation of medical personnel and laboratory procedures which are time-consuming and costly. This study tries to present a machine learning-based approach to predict breast cancer, while adding a classification probability analysis to make the prediction more informative. The breast cancer dataset was used to train four models, namely Logistic Regression, Support Vector Machine, Random Forest, and K-Nearest Neighbor. Evaluation was carried out using accuracy, confusion matrix, ROC curve, and AUC. The results showed that all four models were able to classify cancers with fairly high performance, while one model stood out with the highest accuracy and AUC values. Classification probability analysis provides additional perspective on the confidence level of predictions, which can help medical personnel make more objective clinical decisions.]]>
Copyrights © 2026
