Comparative Approaches by using Machine Learning Algorithms in Breast Cancer Prediction

Abstract

Cancer is most dangerous disease in the world now. It affects the unusual segments of the human body. A sum of Breast cancer patients in the world is second-most among all other cancers. Nowadays it is challenging to detect breast cancer in its early stage. If the researcher gets success to detect cancer in its early stage then it is quite easier to reduce the death rate caused due to this disease. Artificial Neural Network(ANN) can be the solution for automated disease detection in the early stage. Machine Learning (ML) approaches can be the best suit for prediction as these algorithms have intelligent learning ability for improved performance. This work is meant for the prediction of breast cancer in women by using different ML ability-based algorithms and choosing the best performance model. For the total number of 569 data points number of features are included radius, texture, perimeter, area, smoothness, compactness, concavity, and concave points. Here three classification algorithms as Support Vector Classifier (SVC), Random Forest (RF), and XGBoost (XGB) are taken for the performance measure. the classifiers are trained, tested, and validated by taking different parameter settings. In conclusion, XGBoost (XGB) is the best classification in predicting breast cancer using the Kaggle dataset with an accuracy of 98.7%.