Early predictive model for breast cancer classification using blended ensemble learning

Breast cancer is one of the most common cancers among women’s worldwide, and it is a fact that most of the cases are discovered late. Several researchers have examined the prediction of breast cancer. Breast cancer poses a significant hazard to women. The deficiency of reliable predictive models really makes it challenging for clinicians to devise a treatment strategy that will help patients live longer. An automatic illness detection system assists medical personnel in diagnosing disease and provides a reliable, efficient and quick reaction while also lowering the danger of death. A Blended ensemble learning, which is an innovative approach, has been utilized for the classification of breast cancer and this model performs effectively for the base classifier in the prediction analysis. The performance of five machine learning techniques, namely support vector machine, K-nearest neighbors, decision tree Classifier, random forests, and logistic regression, are used as base learners in blended ensemble model. All the incorporated base learners (individually) and the final outcome of the Ensemble Learning are being compared in this study against several performance metrics namely accuracy, recall, precision and f1-score for the early prediction of Breast Cancer. There is a 98.14 percent noticeable improvement with the Ensemble Learning model compared to the basic learners.