Breast Cancer Prediction Using Decision Tree, Random Forest, and KNN Algorithms

Breast cancer (BC) is where abnormal cells grow out of boundaries and convert the tumor to cancer. There are two types of tumors, Benign (B) and Malignant (M). B tumor is a kind of tumor in which cells do not grow out of boundary. However, in M tumors, cells grow automatically and take a form of cancer WHO data shows some facts about BC from the last years that in high-income nations, the age-standardized death rate from BC decreased by 40% between the 1980s and 2020. But in normal-income countries rate of BC is increasing continuously. There is much evidence that pollution is also the reason for BC, whenever women are in existence of polluted environment, the risk of BC increases drastically because the particles of pollution are inhaled and deeply penetrate Breast cells which may cause cancer. But by using Machine Learning (ML) various approaches rate of BC is decreasing, because of early prediction of the disease. In this article, different ML techniques are applied to forecast the BC disease, such as Decision Tree, Random Forest, and K-Nearest Neighbor. These are all supervised learning (SL) techniques and are also called classifiers. Comparison analysis of the models is also discussed in the paper. Out of the three models Random Forest gave the best accuracy of 97%.