A Comprehensive Review on Machine Learning In Breast Cancer Analysis

Abstract

One of the most feared illnesses is cancer since it sneaks up on people and you don't know how to fight it. Early and accurate detection are essential for efficient treatment and better patient outcomes for breast cancer, a critical worldwide health concern. For the investigation, Wisconsin Diagnostic Breast Cancer (Diagnostic) (WDBC) dataset was considered from the repository for machine learning at UCI. Medical records pertaining to breast cancer might produce insightful results using Data Mining techniques, such as trends in behavior and frequent/rare item distribution. To determine the best model, the study has been applied to several Machine Learning classification algorithms -SVM, SMO, NB, Attribute Selected Classifier, Decision Strump, J48, using Weka 3.8.3. In all these algorithms, Wrapper-Subset-Evaluation is used to apply feature selection. On the breast cancer data set, it shows that the J48 method performs better than every other classifier. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.