Diagnosis and Classification of Breast Cancer Using Multiple Machine Learning Algorithms

Abstract

In recent times, Breast Cancer is one of the most commonly diagnosed diseases in women worldwide leading to increased mortality. Cancer is life-threatening and needs early and immediate diagnosis and treatment for the survival of the patient. Early detection will increase the survival rate while reducing the treatment cost of later-stage health complexity. In this paper, we show the use of multiple machine learning algorithms for the diagnosis and classification of breast cancer using the Kaggle Wisconsin Breast Cancer (Diagnosis) Dataset. From our results, we have seen that breast cancer diagnosis using machine learning algorithms such as SVM, and random forest, has achieved greater precision and sensitivity. The performance comparison of the results of various classifiers on the classification shows that the best performance on the accuracy, precision, and sensitivity is achieved with SVM 97% with a precision of about 97% for both benign and malignant cases and a sensitivity of 98% for the benign case and 95% for the malignant case. Consequently, we have seen that machine learning is a good way for an early and faster diagnosis of breast cancer.