K Value Effect on Accuracy Using the K-NN for Heart Failure Dataset

Heart failure is included in the category of cardiovascular disease. disease not easy to detect, and its detection needs be done by experienced skilled medical professionals. Most patients with heart require hospitalization. Common symptoms disease, such as chest pain high or low blood pressure, vary from person person. This study aims find most optimal k value based on accuracy obtained calculations testing different values, namely 1, 3, 5, 7, 9. After getting results five compare which has highest value, best for K-Nearest Neighbor (K-NN) models. The classification process uses K-NN algorithm. algorithm quite use because some parameters work using distance metrics values. Therefore, greatly affects that will produced. In this study, was = 7 9, are they have compared other an 88%. expected benefit research it can make a scientific contribution field machine learning classification, especially predicting