Using Prediction Based Approaches to Improve Diagnostic Accuracy for BreastCancer

Diagnosis BC is a tremendous challenge in the medical domain. In ongoing exploration, the classification of imbalanced dataset certainly stands out enough to be noticed. Imbalanced classification reduces the model's performance by favoring the majority class, although we used a balanced dataset called "Diagnostic Wisconsin BC" in this research. To measure the performance, we compared the different ML algorithms. In order to deploy ML techniques the dataset was divided into train and test data. The hyper tuning settings wereemployed to boost the models' accuracy. Further the ML algorithms compare with the ensemble-based learning techniques like voting classifier, bagging and boosting. Thesecond important part of this article is feature selection for diagnosing BC based on the tumor traits chosen.