Ensemble Learning-based Machine Learning Approach to Predict Breast Cancer

Breast cancer is one of the most lethal forms of cancer, with its mortality rate rising steadily due to global population growth. The disease originates from the uncontrolled proliferation of abnormal cells in the breast, and despite advancements in diagnostic methods, early detection remains a significant challenge because of the microscopic size of cancerous cells at their onset. Early diagnosis is essential for improving treatment outcomes. This research aims to develop machine learning models that predict breast cancer risk by analyzing two distinct datasets: one focusing on external features and the other on the affected breast tissue. The study evaluates various machine learning classifiers and achieves optimal results through ensemble techniques. A voting classifier, combining Random Forest, Logistic Regression, and K-Nearest Neighbors, yielded the highest accuracy for both datasets. Notably, it achieved 99% accuracy for the dataset focused on affected breast tissue and 88% accuracy for the external feature dataset, demonstrating its robustness in predicting breast cancer.