Breast Cancer Prediction using Ensemble Technique

Breast cancer is one the most frequent illness in females, and it is the primary reason why so many women lose their lives to cancer overall. Over the course of the past several years, it has developed into a widespread concern, and its prevalence has increased substantially in recent times. Early identification of breast cancer is the most efficient method for treating its side effects and managing the disease. Women's mortality rates from breast cancer may be lowered thanks to the widespread adoption of Computer-Aided Diagnostic (CAD) devices for finding the disease at an early stage. A large degree of variance is observed when a single model is used. We have proposed ensemble-based models which reduce the variance of the model and hence result in better accuracy. We conducted our experiment on Wisconsin Breast Cancer Database (WBCD) dataset. Experimental results show that the ensemble models are easily outperforming stand-alone models.