Breast Cancer Diagnosis Using K-Nearest Neighbor Algorithm

Breast cancer occurs when cells in the breast grow uncontrollably and abnormally and develop into tumorous tissue. Breast cancer, which is highly prevalent worldwide, can occur in both women and men. It is much more likely to occur in women than in men. Mortality rates from breast cancer provide general information about the health systems of countries. Early diagnosis and treatment are of great importance in breast cancer. Breast cancers are divided into 2 types: benign and malignant. The correct classification of tumors as benign or malignant can prevent patients from undergoing unnecessary treatments. As machine learning methods have developed, their use in health systems has also increased. K-Nearest Neighbor (K-NN) algorithm is a machine learning algorithm used for classification problems. In this study, Wisconsin Breast Cancer dataset is classified using K-NN algorithm. The dataset has a total of 699 samples belonging to 2 classes, benign and malignant. In the study, the effect of 3 different distance metrics and 3 different number of neighborhoods on K-NN performance is evaluated over different metrics. The results are presented in tables and graphs.<br>