Breast cancer is a type of cancer that is commonly formed in breast cells and the cancer cells grow out of control. Cancer can occur in all genders. In 2020, the Global Cancer Observatory recorded a death rate of 684,996,000 and new cases of 2,261,419[1]. From the mortality rate, both men and women should be aware of their health by taking actions such as early detection and avoiding the risk of causing cancer. The source of data in this study came from the UCI Machine Learning Repository. This study aims to compare three data mining algorithms for classifying breast cancer. In this study, the algorithms used in making comparisons are the Decision Tree Algorithm, Naive Bayes, and KNN using 2 cross-validation methods, Hold-Out and K-Fold. The results of the test showed that the KNN algorithm always produced excellent accuracy performance compared to the Naive Bayes and Decision Tree algorithms, namely 98% in the Hold-Out method and 96% in the K-Fold method, while Naive Bayes is 95% on the Hold-Out method and 95% on the K-Fold method, Decision Tree is 94% on the Hold-Out method and 93% on the K-Fold method. Keywords—Breast Cancer, Decision Tree, Naive Bayes, KNN
Copyrights © 2022