Comparative Evaluation of Machine Learning Algorithms in Breast Cancer

Abstract

Breast cancer is one of the world's leading causes of mortality in women and is due to uncontrollable breast cell growth. Early detection and proper care are the only means of avoiding deaths from breast cancer. The precise characterization of tumors is a critical task in the medical profession. Because of their high precision and accuracy, machine learning methods are commonly used in identifying and classifying various forms of cancer. In this review article, the authors have tested different machine learning algorithms and implemented them, which can be used by doctors to identify cancer cells in an early and accurate way. This article introduces several algorithms, including support vector machine (SVM), Nave Bayes Classifier (NBC), artificial neural network (ANN), Random Forest (RF), decision tree (DT), and k-Nearest-Neighbor (KNN). These algorithms are trained with a collection of data containing tumor parameters for a person with breast cancer. After comparing the results, we found the highest accuracy of the Support Vector Machine and Random Forest and the highest precision of the Naive Bayes Classifier (NBC). In addition, we review the number of researches that provide machine learning algorithms for detecting breast cancer