Performance Analysis of XGBoost Ensemble Methods for Survivability with the Classification of Breast Cancer

Breast cancer (BC) disease is the most common and rapidly spreading across globe. This can be prevented if identified early, this eventually reduces death rate. Machine learning (ML) frequently utilized technology in research. Cancer patients benefit from early detection diagnosis. Using machine approaches, research proposes an improved way of detecting breast cancer. To deal with problem imbalanced data class noise, Synthetic Minority Oversampling Technique (SMOTE) has been used. There are two steps suggested task. In first phase, SMOTE to decrease influence imbalance issues, subsequently, next classified using Naive Bayes classifier, decision trees Random Forest, their ensembles. According experimental analysis, XGBoost-Random Forest ensemble classifier outperforms 98.20% accuracy