Efficient Machine Learning-Based Diagnosis System for Breast Cancer

Abstract

AbstractMachine Learning (ML) algorithms find their application in the field of medicine especially in cancer prognosis enabling early diagnosis and treatment. Breast cancer is the predominant form of cancer occurring in women. It is a malignant growth or tumor resulting in uncontrolled division of cells that invades into the adjoining cells and also spreads to different parts of the body. It is the fifth leading cause of cancer mortality. In this paper, diverse ML algorithms are applied to the Wisconsin Diagnostic Breast Cancer (WDBC) dataset. Linear Discriminant Analysis (LDA) and Principal Component Analysis (PCA) are used for feature extraction, while Naive Bayes Classifier (NBC), Random Forest (RF), Neural Networks (NNs) and Support Vector Machine (SVM) are used for classification of data in the WDBC dataset. It is seen that LDA and NNs offer better results in terms of Accuracy, Sensitivity, Specificity and Kappa coefficient.KeywordsMachine learningFeature selectionClassificationBreast cancerBenignMalignant