Article Per Year (5 Year)
p-Index From 2020 - 2025
0.23
P-Index
This Author published in this journals
All Journal Euler : Jurnal Ilmiah Matematika, Sains dan Teknologi
Ilham, Muhammad Rozan Nur
Unknown Affiliation
Computer Science & IT Mathematics

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

Found 1 Documents
Search

 1 
Implementasi Artificial Neural Network (ANN) dalam Memprediksi Nilai Tukar Rupiah terhadap Dolar Amerika Sakti, Adam Indra; Saputra, Lianda; Suhendra, Helen; Halim, Nikken; Alviari, Irfaliani; Ilham, Muhammad Rozan Nur; Putri, Marwah Hotimah Nada; Dalimunthe, Desy Yuliana
Euler : Jurnal Ilmiah Matematika, Sains dan Teknologi Volume 12 Issue 2 December 2024
Publisher : Universitas Negeri Gorontalo

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37905/euler.v12i2.26654

Abstract

The exchange rate of one country's currency against other countries takes an important role in the development and economic activities for a nation. This condition of the Indonesian currency exchange rate, namely the rupiah, is now continuously increasing, meaning that the exchange rate is weakening and experiencing depreciation. Apart from that, the rupiah exchange rate also experiences fluctuations, so forecasting is needed to find solutions to problems that will arise if the currency exchange rate increases. This research purpose is to find the best of network archictecture and to predict the selling rate of the rupiah (Rp) per 1USD for one year. The forecasting method used in this research is using an Artificial Neural Network (ANN) with Backpropagation algorithm. This method is suitable for use in time series analysis because the algorithm is able to adjust the data and has a relatively small error. The data used is the rupiah exchange rate against the USD in the form of time series data, which from March 1, 2019 to February 28, 2024. The data scenario of 90% training and 10% testing at the training stage obtained the best architecture 4-20-1 with MSE is 0.0010385. The data scenario is 80% training and 20% testing where in the training the best architecture is 4-25-1 with an MSE of 0.00089412. The data scenario is 70% training 30% testing where in the training the best architecture is 4-25-1 with an MSE of 0.00099221. Thus, the prediction prices used are predictions for the 80% training data scenario and 20% testing data, because the accuracy results (MSE) are better than the other two scenarios.
Co-Authors Alviari, Irfaliani Desy Yuliana Dalimunthe, Desy Yuliana Halim, Nikken Putri, Marwah Hotimah Nada Sakti, Adam Indra Saputra, Lianda Suhendra, Helen