Claim Missing Document
Check
Articles

Found 1 Documents
Search

Klasifikasi Tingkat Resiko Serangan Penyakit Jantung menggunakan Metode K-Nearest Neighbor Denis Ahmad Ryfai; Nurul Hidayat; Edy Santoso
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 6 No 10 (2022): Oktober 2022
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Heart disease is one of the most common diseases that cause death worldwide. In Indonesia, in 2014 the Sample Registration System (SRS) survey explained that 12.9% of the main cause of death at all ages was coronary heart disease. Early detection of the possibility of heart disease is needed to prevent the worst that can happen to everyone. A classification method that can be implemented into a software for detecting the risk level of heart disease attacks is K-Nearest Neighbor (K-NN), which is a method that classifies based on training data by looking at the closest distance with the Eucledian Distance formula to identify objects class as much as the value of K. Based on analysis and testing in this study, it is known that the results of the training data influence on accuracy ranging from the amount of training data as much as 21 to 210, the value of K = 5 and test data as much as 60 obtained the highest accuracy of 88.333% produced when the amount of training data used is 126. And the results of the analysis on the K values effect on accuracy ranging from K values of 3 to 91, 126 training data and 60 test data obtained the highest accuracy of 96.667% produced when the K values are 57 and 59.