Imbalanced Machine Learning Based Techniques for Breast Cancer Detection
Ahmed Mohammed, N. Arunachalam · 2021 International Conference on System, Computation, Automation and Networking (ICSCAN) · 2021
Breast cancer is the world's second-leading cause of cancer death among women. It is the most frequent type of invasive cancer in women and the second biggest cause of mortality in the United States. Breast cancer is a disease in which the breast cells develop out of control. Many studies have been conducted in this area utilizing a machine learning technique, however some of them did not explore how to cope with imbalanced classes. The breast cancer dataset is an unbalanced dataset in which the total frequencies of benign class is larger than the total frequencies of malignant class. The problem with an unbalanced dataset is that it provides a high level of accuracy just by predicting the majority class, but fails to capture the minority class, which is usually the point of creating the model in the first place. Imbalanced classification approaches were employed in this study to predict breast cancer. The accuracy of prediction for multiple machine learning algorithms was determined utilizing imbalanced methodologies such as synthetic minority oversampling technique, random oversampling, and random under sampling technique respectively.