Machine Learning Algorithms for Binary Classification of Breast Cancer

Abstract

Breast cancer is one of the most serious diseases in women, and it is the second largest cause of cancer mortality. Breast cancer occurs when malignant, cancerous tumors form from breast cells. Self-tests and routine clinical examinations aid in early detection and, as a result, dramatically improve survival chances. Breast cancer classification is a medical strategy that researchers and scientists face great challenges. Neural networks have acquired appeal as a technique for cancer data classification. This research begins by looking into the use of machine learning algorithms to classify breast masses for various diagnostic, predictive, or prognostic tasks in a number of imaging modalities, including MRI, ultrasound, digital breast tomosynthesis, and mammography. For evaluation, we proposed various machine learning algorithms such as support vector machine, random forest, Naive Bayes, logistic regression, K-nearest neighbor, artificial neural network, and deep neural network. We have used publicly accessible Breast Cancer Wisconsin (Diagnostic) datasets from the UCI Repository. As demonstrated in the results sections, the DNN classifier has a high level of accuracy (99.42%), showing that it outperforms traditional models.