Claim Missing Document
Check
Articles

Found 1 Documents
Search

Optimasi Parameter Support Vector Regression Dengan Algoritme Genetika Untuk Prediksi Harga Emas Muthia Azzahra; Budi Darma Setiawan; Putra Pandu Adikara
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 2 No 1 (2018): Januari 2018
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (1068.621 KB)

Abstract

Gold is one of the precious metals that many people interested as commodity to invest because of its resistance to inflation. Fluctuations can occur so extreme that affect the value of gold. Therefore, prospect of gold value in the future is quite important for the investors. One of prediction methods is Support Vector Regression (SVR), but the sensitivity of SVR parameters could influence the prediction result, therefore Genetics Algorithm (GA) can be applied, this method is flexible enough to be hybridized. This study discuss about the optimization of SVR parameters using GA to predict gold prices. Based on the testing result, the best mean absolute percentage error (MAPE) is 0.2407% with SVR loop 50, GA's generation 95, population size 70, crossover rate 0.01, mutation rate 0.99, elitism percentange 80%, range of 1x10-7-1x10-4, range of 0.01-5, range of 1x10-7-1x10-4, range of 1x10-5-1x10-4, and range of 1x10-3-0.1.