Breast Cancer Prediction Feature Selection Using ML Algorithms

Abstract

Abstract Breast cancer is a disease that breast epithelial cells uncontrolled hyperplasia under the effects of several different carcinogens. The early symptoms of breast cancer are breast lesions, nipple discharge and axillary fossa lymphadenectasis. Breast cancer can directly threat the life of patient by casing various organ lesions and metastasis of cancer cells in the late stage of suffering from this disease. The method of classifying tumors into malignant or benign is the main challenge that needed to be dealt with to treat patient in a correct way. This paper shows the different classification models to predict the kind of breast cancer. All data used to distinguish the type of breast cancer in the research comes from UCI repository. Some simple classifiers are employed in the research, for instance, Decision Tree Classifier, Extreme Gradient Boosting, Naïve Bayes Classifier, Gradient Boosting, Support Vector Machine. The results show that using Support Vector Machine algorithm the prediction accuracy of breast cancer reaches 99.7%.