Optimized Feature Selection for Enhanced Breast Cancer Detection Using Machine Learning

The fundamental causes and preventive strategies of breast cancer are still unknown, early recognition is crucial to increasing the disease’s survival chances. This study investigates the application of machine learning for breast cancer diagnosis using six standard and customized classification models. With the use of four feature selection techniques applied to the Wisconsin Diagnostic Breast Cancer (WDBC) dataset, the Random Forest (RF) model yielded the best accuracy of 98.25%. Integrating feature selection with RF boosted diagnostic performance, demonstrating the potential to eliminate errors and improve treatment planning for clinical applications.