Predicting Breast Cancer An Evaluation of Machine Learning Approaches

Abstract

A disease in which the cells of the breast have uncontrollable growth or cancer cell growth in the breast is known as breast-related cancer. There are various types of uncontrollable growth. In women, it is most common cancer which accounts for around 25 percent of all cancers diagnosed in women. According to the 3rd of February 2023, departmental news of WHO, every year, more than 2.3 million new instances of breast cancer are identified. Breast cancer is one of the major causes of female deaths in 95% of countries. Early diagnosis of BC can decrease the death rate drastically as it is a curable disease when diagnosed in early stages. The classification of diagnosed breast cancer into malignant or benign tumor is a major field of research at present. Use of ML in diagnosis and classification of breast cancer is widely recognized due to its powerful feature in detection from BC datasets. This paper will evaluate various ML techniques such as regression, support vector machines (SVMs), DTs naive Bayes and random forest in the classification and diagnosis of breast cancer. We have used Wisconsin breast cancer diagnosis (WBCD) dataset for paper. A health care system model is also presented along with this paper.