Revised grey Wolf optimized SVM-KNN ensemble based automated diagnosis of breast cancer

Abstract

Breast cancer causes a high percentage of mortality among women throughout the world. For improving breast cancer survival and results, early detection is important. The primary aim of the given study is to enhance the results of automated expert system (ES) in the field of medical diagnosis and come up with more accurate diagnosis system for breast cancer detection at early stage. This research introduces a new ensemble approach that accurately diagnose the breast cancer at a pre-developed stage. The recognized Wisconsin Diagnosis Breast Cancer (WDBC) dataset is used to provide raw breast cancer data in our study. Our proposed model rGWO-KSE (revised Grey Wolf Optimized SVM KNN Ensemble) implements an SVM-KNN ensemble and optimizes it with our newly introduced rGWO (revised Grey Wolf Optimization) technique. The performance of the proposed algorithm is evaluated based on several measures, e.g. accuracy, sensitivity, specificity, f-score and precision. Our model achieves 98.83 % accuracy in detection of breast cancer on WDBC dataset. The comparison with the previous studies for breast cancer detection proves that the proposed study provides better values for performance measures.