Article Per Year (5 Year)
p-Index From 2020 - 2025
0.23
P-Index
This Author published in this journals
All Journal Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer
Dhatu Kertayuga
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
Journal : Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer

 1 
Prediksi Nilai Ekspor Impor Migas Dan Non-Migas Indonesia Menggunakan Extreme Learning Machine (ELM) Dhatu Kertayuga; Edy Santoso; Nurul Hidayat
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 5 No 6 (2021): Juni 2021
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Indonesia's resource wealth is one of the important assets for a developing country. To advance the wheels of the Indonesian economy, trade activities between countries are carried out, namely exports and imports. Resources exported and imported by Indonesia are oil and gas and non-oil and gas resources. Although Indonesia is capable of producing its own oil and gas and non-oil and gas products, Indonesia's imports of oil and gas and non-oil and gas are still higher than Indonesia's total exports of oil and gas and non-oil and gas. To assist Indonesia's economic development strategy, a prediction is needed to estimate the value of Indonesia's oil and gas and non-oil and gas exports and imports. In this study, the algorithm used is Extreme Learning Machine (ELM). Then, the data used are oil and gas and non-oil and gas export data as well as oil and gas and non-oil and gas import data obtained from the Badan Pusat Statistik (BPS) from January 1993 to December 2020. The results obtained from this study are export data with the average mean absolute percentage error (MAPE) value of 6.6742% for the comparison of the number of training : testing, the number of data features, and the number of hidden neurons the best is 70%:30%, 5, and 8. While for import datasets, the comparison of the number of training : testing, the number of data features, and the number of hidden neurons is the best 80%:20%, 4, and 10 with a final MAPE average of 10.0515%.
Co-Authors Edy Santoso Nurul Hidayat