Klasifikasi Status Hipertensi Pasien UPTD Puskesmas Sempaja, Kota Samarinda Menggunakan Metode K-Nearest Neighbor

Abstract

Data mining is a method of selecting, exploring and modeling large amount of data to find knowledge and clear patterns or interesting relation of the data and useful in the process of data analysis. In data mining there are several techniques that have different function and one of them is classification tehcnique. The classification process itself is the process of finding patterns or differences between classes or data that can be used to predict object classes whose class labels are unknown. K-nearest neighbor (K-NN) is one of the methods in classification algorithm. This study discusses the classification using K-NN algorithm which is applied to the data hypertension status. The aim is to find out the optimal neighborliness value (K) accuracy value and the best propotion of the data hypertension status. The data used is the data of patients UPTD health center Sempaja, Samarinda city from February to May 2022 with dependent variabel is hypertension status and uses 4 independent variables, age, gender, diabetes mellitus and heart disease. Based on the research that has been done, obtained an accuracy value of 62,60% with K = 5 in the best proportion of the data is 70%:30%.