A Statistical Classification of the Benign and Malignant Neoplasm using Ensemble Learning and Classification Algorithms

Detection of Breast Cancer is the introductory stage of a cancer diagnosis. Therefore, it is essential to identify whether the tumor is cancerous or non-cancerous. The goal of the paper is to recognize the benignancy and malignancy of the cancerous cell precisely. The paper examines two machine learning algorithms and proposes a novel ensemble learning approach using a blend of classification algorithms namely ℌk-nearest neighbors (k-NN) with Neighborhood components analysis, and Support-vector machine classifier (SVM)ℍ. The proposed unification was examined on the Wisconsin Diagnostic Breast Cancer dataset, provided on the DCI machine learning repository. The suggested approach exhibits a better performance of 99.29% (10- fold CV), 98.41 % (10- fold CV), and 98.85% (10-fold CV) for k-nearest neighbors (NCA), Support-vector machine, and Ensemble method respectively.