Breast cancer diagnosis and survival prediction using ML algorithms

Abstract

Breast cancer is accounted for to be the most well-known malignancy type among ladies worldwide and it is the second most elevated lady’s casualty rate among all malignant growth types. Precisely anticipating the endurance pace of bosom disease patients is a significant issue for malignancy scientists. Machine Learning (ML) has drawn in much consideration with the expectation that it could give exact outcomes, yet its displaying techniques and forecast execution stay dubious. This paper centres on the use of AI calculations for anticipating Haberman's Breast Cancer Survival analysis. Various AI approaches specifically Decision tree, Multilayer Perceptron (MLP), Support Vector Machine (SVM) and K Nearest Neighbour (KNN) strategies are considered for the conclusion of Breast Cancer Survival anomaly. The presentation of arrangement of strange and typical Breast Cancer Survival patients is assessed as far as various variables including preparing and testing exactness, accuracy and review. The point of this deliberate survey is to recognize and basically assess current examinations with respect to the use of ML in foreseeing the 5-year endurance pace of bosom malignant growth. Test results on Haberman's Breast Cancer Survival dataset show the predominance of MLP proposed technique by coming to 96.7% as far as precision.