Penggunaan Algoritma C4.5 dan Random Forest guna Meningkatkan Efisiensi Klasifikasi Penyakit Stroke

Abstract

Stroke is a very serious problem throughout the world. According to a report from the World Heart Organization (WHO), in 2022, more than 12.2 million, or one in four people aged 25 years will experience a stroke, and more than 7.6 million new stroke sufferers every year throughout the world. An irregular lifestyle is the main cause of someone having a stroke. Therefore, we need a system that can be used as a stroke classification or detection tool based on a person's disease history. The stroke disease data used in this study was obtained through Kaggle with a total of 5110 data.  Based on the results of research that has been carried out using two algorithm models, namely the C4.5 algorithm and Random Forest. A combination of these two algorithms has been obtained which produces a stroke classification system with fairly good accuracy, with an accuracy value of 92.4%.