Breast Cancer Prognosis using Machine Learning Techniques: A Literature Survey

Globally, breast cancer (BC) is women's leading cause of death. Primarily breast tissues are involved in the development of breast cancer. The population growth in medical research has made it increasingly necessary to study cancer in its early stages. A growing population has resulted in an exponential increase in breast cancer deaths. Among women, it is the second most common cancer. Cancer prediction is crucial for updating treatment aspects and patient survival standards. Aside from being a research center and a good strategy, machine learning (ML) methods are particularly effective in predicting breast cancer and detecting it early. The Wisconsin Cancer Diagnostic dataset (WBCD) was used to compare five ML algorithms: decision trees (C4.5), support vector machines (SVMs), logistic regression (LR), random forests, and convolutional neural networks (CNNs). Through machine-learning algorithms, this study aims to identify which algorithms are best suited for the early detection of breast cancer and the prediction and diagnosis of the disease using predictive algorithms. As a result of this study, different machine learning algorithms were compared, and it was found that Random Forest had higher accuracy, with a 99.8% accuracy rate.