A Novel Hybrid Machine Learning Framework for Breast Cancer Prediction and Classification

Breast cancer (BC) is a common and deadly health illness that affects women throughout their lives. Therefore, a vital aspect in both enhancing survival outcomes and getting the best treatment results is the early discovery of conditions and correct medical diagnosis. Further, changes in breast cancer prediction accuracy rates will be evaluated between Random Forest (RF) alongside Support Vector Machine (SVM), K-Nearest Neighbors (KNN), Decision Tree (DT), Logistic Regression (LR), and Naïve Bayes (NB). Additionally, the research presents a hybrid model that integrates RF, KNN, and NB components to boost the classification outcomes. Models are assessed using primary measurement criteria made up of accuracy, precision, and recall, together with the F1-score. Moreover, models are assessed using primary measurement criteria made up of accuracy, precision, and recall together with the F1-score. However, the hybrid model achieves higher accuracy levels of 98.2% that exceed both SVM (97.0%) and RF (97.4%). The outcomes demonstrate that ensemble approaches are useful for enhancing prediction accuracy and lowering categorization errors. The research demonstrates the need to merge diverse algorithms into BC diagnostic systems, which leads to better patient outcomes through early detection.