An Empirical Investigation into Machine Learning Approaches for Breast Cancer Classification

Breast cancer was diagnosed in 2.3 million individuals worldwide in 2022, and was the cause of 670,000 deaths. It ranks as the second most common cancer overall and the most frequently diagnosed cancer in women. Since survival is highly dependent on early detection, the development of computational models for diagnosis is crucial. Among the various models available, we have chosen to focus on three supervised machine learning models in this study: Support Vector Machine (SVM), K-Nearest Neighbor (KNN), and Feed-forward Neural Networks (FFNN). Their roles in breast cancer detection are investigated. We have pre-processed the data using normalization and feature selection to enhance these models’ performances. They are trained on real-life documented patient data to predict breast cancer. It is determined that the FFNN is more accurate than both SVM and KNN. This paper provides a comprehensive analysis of the various machine learning models, highlighting their positive and negative aspects in diagnosing breast cancer. The results underscore the potential of advanced neural networks in medical diagnostics and pave the way for further research and development in this domain.