PENGELOMPOKAN TINGKAT RESIKO PENYAKIT JANTUNG BERDASARKAN USIA MENGGUNAKAN ALGORITMA K-MEANS

Abstract

Heart disease is one of the non-communicable diseases that can cause death, This disease occurs due to a narrowing of the blood vessels so as to cause impaired heart function Some of the causes of heart disease are one of them based on age, basically heart disease can be prevented by various factors including a healthy lifestyle, besides that early detection of heart disease is also needed to prevent death in sufferers One way to do early detection is to use data mining. The use of the k-means algorithm can be done to cluster the grouping of heart diseases by age to find out someone is exposed to the cause of high and low heart disease. Based on these problems, this study uses the CRISP-DM (Cross-Industry Standard Process for Data Mining) method with several stages such as Business Understanding, Data Understanding, Data preparation, Modeling, Evaluation, and Deployment. The clustering method with the k-means algorithm in this study shows a new insight, namely grouping the risk level of heart disease based on 3 clusters. Cluster 0 is an age category with a fairly low risk level of heart disease or Low, which is 355 out of 1025 age categories tested, then cluster 1 is an age category with a moderate or Medium heart disease risk level, which is 208 out of 1025 age categories tested, and finally cluster 2 is an age category with a fairly high age category or High, which is 462 out of 1025 age categories tested.