Breast Cancer Prediction Using Unsupervised Learning Technique K-Means Clustering Algorithm

Abstract

Breast cancer is a highly deadly and common cancer that is spreading over the world and claiming many lives. Breast cancer develops in the glandular tissue of the breast in the lining of epithelial cells of ducts or (15%) lobules. These tissues heal with time. These in situ tumours may develop over time and infect the breast cells' immediate environs (stage 1), impact the lymph nodes, and finally totally spread throughout the body's organs (stage 3). There were 685,000 deaths worldwide in 2020 and a later rise in the number of malignant women. Cancer was a common disease that affected 7.8 million of the female population. The k-means algorithm is a popular data clustering algorithm. As the main analytical routine in data mining, the techniques of the clustering algorithm will impact the clustering outcome directly. This paper examines the shortcomings of the standard k-means algorithm and discusses them. This paper reviews existing methods for selecting the number of clusters for the algorithm. However, one of its drawbacks is the requirement that the number of clusters, K, be specified before the algorithm is applied. Therefore, we obtained an accuracy of 85% after executing the program, and we were able to depict the difference between a benign and a malignant tumour. And we were able to see the centroid between the two tumours.