Breast Cancer Prediction Using Different Machine Learning Algorithms: A Comparative Study

Abstract

According to the World Health organization’s (WHO) 2020 report, 2.3 million new cases of breast cancer were recorded, and 685,000 women died due to breast cancer. To treat breast cancer early, a lot of research has been proposed using different types of techniques in the past few years. In recent years, machine learning algorithms (MLAs) have become popular for detection due to their improved accuracy and performance. This paper used 13 supervised machine learning (SML) techniques, namely: Decision Tree (DT), Logistic Regression (LR), Random Forest (RF), Naive Bayes (NB), K-Nearest Neighbors (KNN), Support Vector Machine (SVM), Gradient Boosting (GB), Extreme Gradient Boosting (XGB), Adaptive Boosting (AB), Categorical Boosting (CB), Light Gradient Boosting Machine (LGBM), Multi-Layer Perceptron (MLP), and Extra trees (ET) to predict the outcomes of the Wisconsin Breast Cancer Original (WBCO) dataset from the UCI repository. When all thirteen algorithms were evaluated and compared, MLP outperforms them all with the highest accuracy (98.76%). This accuracy value is 0.56% greater than the recently reported accuracy value of 98.2% for the MLP classifier for the same dataset.