PREDIKSI HARGA EMAS MENGGUNAKAN METODE ARTIFICIAL NEURAL NETWORK BACKPROPAGATION
M. Hadiyan Amaly, Hanipar Mahyulis Sastriana, Wiwit Pura Nurmayanti · Jurnal Eksbar · 2025
The price of gold which continues to increase makes business people or people often make gold for investment. That is why investors must be able to predict the price of gold with the aim of knowing the investment opportunities for gold prices in the future, so that it can be a consideration for gold investors in making investments. The forecasting method used in this research is Artificial Neural Network Backpropagation. The results of this study show that the gold price forecasting architecture model is 3-5-3-1 where there are 3 input layer variables, 5 first hidden layers, 3 second hiden layers, and 1 output value. Producing a prediction of 1812,473 in addition to the MSE value of 0.0011034 and the accuracy value of the model using MAPE of 10.71607% so that the forecasting model is good and accurate.