A Review of Machine Learning Techniques for Breast Cancer Diagnosis in Medical Applications

Abstract

During their life, among 8% of ladies are determined to have Breast cancer (BC), after Lung Cancer. Breast cancer is the second predominant cause for the increasing fatal rate. Breast cancer is exposed by the transformation of qualities, consistent agony, changes in the size, colour (redness), and the skin surface of the breast. Classification of breast cancer leads the pathologists to discover a specific and target prognostic, by binary classification (normal / abnormal). This work associates five of the most prevalent machine learning methods utilized for breast cancer exposure and analysis, explicitly Support Vector Machine (SVM) for breast cancer exposure and analysis, explicitly S.V.M, k-Nearest neighbours, logistic regression, decision tree, and random forest. The Wisconsin dataset is utilized to estimate and compare the accuracy of five different ML methods with the evaluation metric viz. accuracy. The best performance is obtained by decision tree and random forest with the highest accuracy.