Prediction of Breast Cancer Using Ensemble Learning

Abstract

Among all types of cancers, breast cancer is the most crucial and fatal cancer particularly for women as it is considered the second leading cause for cancer death among women. When it’s a question of survival, correct treatment is a vital requirement, and for proper treatment, the first requirement is to identify cancer accurately. Here, we are motivated to provide a highly reliant and consistent system for the prediction of breast cancer. In the proposed method, we have used ensemble learning for the desired accuracy. The ensemble voting system comprises a total of five machine learning (ML) classifiers which include Random Forest, Naive Bayes, SVM with two different kernels (rbf, polynomial), K-Nearest Neighbors and Decision Tree. We experimented on the Wisconsin Breast Cancer Dataset from UCI machine learning repository. We achieved a maximum testing accuracy of 99.28% and a maximum precision of 97.22% while using 5-fold cross-validation. The proposed system exhibited satisfying accuracy on the dataset and outperformed many of the prominent existing methods.