Predictive Performance Analysis of Ensemble Learners on BCD Dataset
Gotam Singh Lalotra, Vinod Kumar, Dharmendra Singh Rajput · 2021
Breast cancer disease (BCD) has now become a worldwide disease in human beings. The breast cancer cases are progressively rising especially in females. Initial detection of this disease can save a patient’s life and cost of treatment. There are many clinical diagnosis methods which are used for the detection of breast cancer. Among them supervised machine learning is a very powerful mechanism that helps in predicting the presence or absence of disease based on the laboratory test data of the patient. Presently, numerous machine learning algorithms are available for classifying the disease in patients, but the best amongst all is yet to be explored. Therefore, this research work contributes to find the outperforming model from six non-ensemble methods and four ensemble methods for breast cancer detection. This study claims AdaBoost the most stable and suitable model for BCD prediction with accuracy 97.1%, precision, recall, F1-measure and AUC 95.91 %.