Diagnosis of Breast Cancer Using Entropy Weight-Feature Selection and Ensemble Learning

The discovery of huge medical databases with the help of new computational tools has confirmed the existence of all diseases, including certain cancers. However, medical research now tends to look at diseases separately from each other, rather than focusing on their interactions. So far, various methods based on data mining and machine learning techniques have been proposed, which, despite their many applications in the field of diagnosis of breast cancer, still face the problem of insufficient accuracy, which has become a major challenge. Therefore, in this paper, for the diagnosis of breast cancer using the entropy-based feature selection algorithm, the weight-prominent feature is selected and then the ensemble learning system consisting of SVM, KNN, Adaboost and ID3 decision tree machine learning algorithms used. In this paper, the SEER dataset has been used. After simulating the proposed method, it was observed that the accuracy of the proposed method on average compared to other methods such as SVM, KNN, Adaboost and ID3 decision tree on breast cancer data has improved by about 11.5% compared to other methods.