On Predicting and Analyzing Breast Cancer using Data Mining Approach

Abstract

The highest invading cancer among the women is breast cancer. Early detection of breast cancer is the higher chance of the patient being treated. In this study, we have proposed an ensemble method named stacking classifier which combines multiple classification techniques and efficaciously classifies the benign and malignant tumor. “Wisconsin Diagnosis Breast Cancer” dataset culled from the UC Irvine Machine Learning Repository has been used for our experiment. We applied different classification techniques over the dataset and tuned their parameters to improve accuracy. We chose the three best classifiers for our proposed method. Generally, our proposed Stacking classifier combined the results of those best classifiers using meta classifier and provided 97.20% accuracy for breast cancer prediction. Performance of different data mining approaches have been evaluated rigorously through different evaluation metrics.