Jurnal DISPROTEK
Vol 10, No 2 (2019)

PERBANDINGAN METODE ANN-PSO DAN ANN-GA UNTUK PENINGKATAN AKURASI PREDIKSI HARGA EMAS ANTAM

Nadia Annisa Maori (Universitas Islam Nahdlatul Ulama Jepara)



Article Info

Publish Date
14 Oct 2021

Abstract

Antam's gold price is more expensive than the price of gold used by more investors for long-term use. Sometimes the price of antam gold cannot be predicted at any time. Antam's rising gold prices were moved by many factors, sent in exchange rates of US dollars (USD). If the exchange rate of the US dollar (USD) decreases, the price of gold will rise and vice versa, if the value of the US dollar (USD) strengthens, the price of gold will increase. This condition makes it difficult for investors to predict the price of gold in the future. Backpropagation Artificial Neural Network (ANN) is known as one of the good methods in predicting. In this study an evaluation of the results of the price of gold using ANN with the help of PSO (Particle Swarm Optimization) and GA (Genetic Algorithm) optimization. PSO has many similarities with GA, which is an algorithm adopted from the process of supporting humans. The results of the study prove that PSO Optimization is able to provide an increase in optimizing the weights on the Neural Network by producing the best RMSE value, which is equal to 0.026, while GA optimization only produces a value of 0.09.

Copyrights © 2019






Journal Info

Abbrev

JDPT

Publisher

Subject

Civil Engineering, Building, Construction & Architecture Computer Science & IT Electrical & Electronics Engineering Industrial & Manufacturing Engineering Other

Description

Jurnal DISPROTEK e-ISSN 2548-4168 p-ISSN 2088-6500 Jurnal di publikasikan oleh Fakultas Sains dan Teknologi, Universitas Islam Nahdlatul Ulama Cakupan isi jurnal DISPROTEK dalam bidang teknik adalah : teknik elektro, teknik sipil, teknik industri; dalam bidang ilmu komputer adalah : teknik ...