Breast Cancer Classification using the Supervised Learning Algorithms

Abstract

Breast cancer (BC) has been the second largest cause of death for women around the world for the past few years. BC is characterized by the chronic pain, genes mutation, color (redness), changes in the size and texture of the skin. BC classification helps clinicians to find a comprehensive and accurate response to treatment, with the most common binary classification (benign / malignant cancer). Nowadays, the Machine Learning (ML) techniques are commonly used in the case of classification of breast cancer. They support with high classification accuracy and rapid evaluation technologies. The proposed research work is mainly focused on supervised learning algorithm, which uses four distinct classifiers: K-Nearest Neighbor (KNN), Weighted K-Nearest Neighbor (WKNN), Support Vector Machine (SVM), Linear Discriminant Analysis (LDA) and Artificial Neural Network (ANN) for the classification of breast cancer. Also, this research work suggests the difference between the aforementioned classifiers and determines their accuracy. The performance of the classifier is assessed based on its accuracy, sensitivity, specificity, precision and recall. Results indicate that, ANN provides the highest accuracy of 97.60% than the other classifiers.