IMPLEMENTASI DATA MINING UNTUK MEMPREDIKSI PENYAKIT JANTUNG MENGGUNAKAN METODE K-NEAREST NEIGHBOR DAN LOGISTIC REGRESSION

Abstract

Heart attack disease is a condition where the arteries are blocked due to fatty deposits. This disease causes several symptoms such as shortness of breath, chest pain. In addition, this is also due to impaired blood flow to the heart that is blocked and can destroy the heart muscle. Until now, heart attack disease is still the leading cause of death in Indonesia. The problem faced today is that it is very difficult to predict heart disease and determine whether a person has heart disease. An appropriate method is needed to predict heart disease. The purpose of this study was to calculate the level of accuracy in predicting heart attack using the K-Nearest Neighbor and Logistic Regression methods. Based on the research and data processing that has been applied and the comparison of the K-Nearest Neighbor and Logistic Regression algorithms, the final results are the accuracy of the Logistic Regression Algorithm of 88% and the K-Nearest Neighbor algorithm of 83%. Thus it can be concluded that the Logistic Regression algorithm is the best in predicting heart attack disease than the K-Nearest Neighbor algorithm.