Perbandingan Metode Euclidean Distance dan Haversine Distance pada Aplikasi Sistem PPDB dan algoritma K-Means Untuk Menentukan Kebijakan Peraturan Zonasi

Abstract

At this time, registration for public schools is straightforward to do online with a device via a web browser without the need to install an application that can ease the device's performance. Still, the government regulates it through a zoning policy that makes students register for schools close to their homes. This study examines and compares which Euclidean and Haversine algorithms are more accurate to implement in making an application that determines the distance between the school and the student's house. Then the school will decide which students can be accepted using the K-Means algorithm, as has been done by SMPN 1 Tigaraksa, which results that the haversine algorithm has an average accuracy rate of 99.71%, an average error of 0.29% with an average distance difference of 1.86 meters. In comparison, Euclidean has an accuracy rate of 99.65%, an average -the average error is 0.35% with the difference in average distance at the actual length of 2.42 meters. Therefore, the difference in distance between the two algorithms obtained is 1.27 meters. And the K-Means Algorithm can be the proper method for making decisions because the algorithm groups according to the farthest, medium, and closest distances