Article Per Year (5 Year)
p-Index From 2020 - 2025
0
P-Index
This Author published in this journals
All Journal Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer
Rezza Hary Dwi Satriya
Fakultas Ilmu Komputer, Universitas Brawijaya
Computer Science & IT Control & Systems Engineering Education Electrical & Electronics Engineering Engineering

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
Implementasi Metode Ensemble K-Nearest Neighbor untuk Prediksi Nilai Tukar Rupiah Terhadap Dollar Amerika Rezza Hary Dwi Satriya; Edy Santoso; Sutrisno Sutrisno
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 2 No 4 (2018): April 2018
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (1368.433 KB)

Abstract

The exchange rate is the currency unit price agreed by each country as a means of payment or transaction. The most used exchange rate in Indonesia is the rupiah exchange rate against the dollar. The dollar is the most stable currency in the economy. The high or low of the rupiah exchange rate is influenced by rates of interest, inflation, exports, imports, and sovereign debt. The exchange rate also has an important role in determining economic policy. In order to obtain an appropriate economic policy in the future situation and conditions, it is necessary to use a solution by using Ensemble kNN algorithm to predict the future rupiah exchange rate. The count of data was used in this research are 24 data training and 12 data testing. The data training and testing consists of 5 parameters, such as BI rate, Inflation, Export, Import, and sovereign debt. The Ensemble kNN algorithm uses a supervised learning, which the data testing is classified based on the majority of classes on kNN. The principle of kNN is to find the K variable from the data training which having closest similarity to the data testing. Ensemble technique is used to optimize kNN algorithm to get more accurate result. The result from this prediction system was evaluated by using MAE, MAPE and RMSEP. The obtained value of MAE buy = 456.56, selling MAE = 460.96, MAPE buy = 3.47%, MAPE selling = 3.47%, and RMSEP buy = 534.88, RMSEP selling = 540.07. The final result is the conformity of result and the pattern which produced between the predicted data and the actual data.
Co-Authors Edy Santoso Sutrisno Sutrisno