Data Mining untuk Klasifikasi Penderita Kanker Payudara Berdasarkan Data dari University Medical Center Menggunakan Algoritma Naïve Bayes

Abstract

The breast cancer sufferers data from the University Medical Center is data about patients suffering from breast cancer based on certain characteristics. This data has abundant information so that data mining can be done with the aim of digging deeper information which of course can be useful in the future. The data class itself is divided into 2 groups: recurrence and non-relapse classes. The technique used in this study is classification using the Naive Bayes algorithm. Naive Bayes is a simple probabilistic prediction technique based on the implementation of Bayes rules with a strong assumption of independence on features. The tool used to find accuracy values is RapidMiner 9.3. Data attributes consist of Class, Age, Menopause, Tumor-Size, Inv-Nodes, Node-Caps, Deg-Malig, Breast, Breast-Quad and Irradiant. In terms of methods, this study uses the CRISP-DM (Cross Industry Standard Process for Data Mining) method. This research is used as information in making decisions to determine policies taken in dealing with patients with breast cancer