PBC-ML: Predicting Breast Cancer in Humans using Machine Learning Approach

Abstract

Cancer is a disease in which cells grow uncontrollably, potentially causing harm tosurrounding healthy tissue and organs. Breast cancer is a specific type of cancer that affectsthe breast and is the second most common cancer among women worldwide. Symptoms ofbreast cancer include a lump or tumour, swelling, nipple discharge, and swollen lymphnodes. Breast cancer is staged, with stage 0 being the earliest stage with minimal symptomsand stage 4 indicating the cancer has spread to other parts of the body. The future burdenof breast cancer is predicted to increase, with over 3 million new cases and 1 million deathsin 2040. Early detection is crucial for successful treatment and recovery, and machinelearning can be used to predict the likelihood of breast cancer based on symptoms. So wepropose in our research to use machine learning algorithms such as CART, SVM, NB, andKNN to analyse and build models for breast cancer detection. These findings offer asummary of relevant machine learning methods for breast cancer detection as it will help tocurb it and we got an accuracy of 98.2% compared to the state of art methods which hasaccuracy of 99%. It proves to be a valuable tool in the early detection of breast cancer andcan improve the accuracy of existing diagnostic methods.