Prediksi Persentase Penyelesaian Permohonan Hak Milik Menggunakan Metode Support Vector Regression (SVR) (Studi Kasus : Badan Pertanahan Nasional (BPN) Kabupaten Malang)

Wa Ode May Zhara Averina (Fakultas Ilmu Komputer, Universitas Brawijaya)
Edy Santoso (Fakultas Ilmu Komputer, Universitas Brawijaya)
Novanto Yudistira (Fakultas Ilmu Komputer, Universitas Brawijaya)

Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer
Vol 4 No 9 (2020): September 2020

Ownership Certificate (SHM - Sertifikat Hak Milik) is one type of certificate in which the owner has a full right to the land ownership in a certain area that has been mentioned in the certificate itself. Apparently, the service of community land certification takes a long time resulting the number of SHM that can be completed each month is lower than the number of applications that enter. This resulted a buildup of management by officers from time to time. To minimize this, a research needed to be conducted using a method of predicting, one of them was by using the Support Vector Regression method that is able to solve regression problems and produce good performance in taking solutions. Based on the results of tests conducted for this SVR method, the predicted results with a minimum evaluation value using MAPE were 0.3308% with lambda parameter values = 0.027, sigma = 0.01, epsilon = 0.00002, cLR = 0.24, C=100, and the number of iterations=10. The MAPE value obtained was <10%, including the number of documents in the calculation process.