Breast Cancer Detection Using Ensemble Learning Model

One of the leading causes of mortality rate in women nowadays in worldwide is cancer, especially breast cancer. Historically, clinical screening was used to make a diagnosis, which was then followed by thorough family genetic examination. Breast cancer is regarded as a big concern due to the number of fatalities it causes because no vaccination has been invented yet, but thanks to modern medical science's advancements, it is fully curable if caught in the early stages without harming the patient. The main difficulty relied in appropriate diagnosis, of any sort of cancer in the patient, whether it is malignant or benign. Reliable and cost-effective techniques used in Machine learning and deep learning might be helpful in these predictions rather than traditional models. Early disclosure and determination could gain the possibilities of fruitful treatment and life span of the ladies. In this paper, the Wisconsin breast cancer diagnostic data set which was gathered from the UCI machine learning repository was used to prepare and test the models. Proposed algorithm was compared with Logistic regression (LR), support vector machine (SVM), Decision tree (DT), K-Nearest Neighbor (KNN), and Random Forest (RF). The proposed ensemble XG Boost learning model is contrasted with the traditional algorithms and proven that the model gives accurate results with a value of 97%.