Implementation of LSTM-RNN for Bitcoin Prediction

Nur Ghaniaviyanto Ramadhan (Institut Teknologi Telkom Purwokerto)
Nia Annisa Ferani Tanjung (Institut Teknologi Telkom Purwokerto)
Faisal Dharma Adhinata (Institut Teknologi Telkom Purwokerto)

Indonesian Journal on Computing (Indo-JC)
Vol. 6 No. 3 (2021): December, 2021

Bitcoin is a cryptocurrency that is used worldwide for digital payments or simply for investment purposes. Bitcoin is a new technology so there are currently very few prices prediction models available. Problems arise when someone uses bitcoin without understanding strong fundamentals. This can result in a lot of loss for the person. These problems certainly need to be overcome by predicting bitcoin prices using a machine learning approach. The purpose of this research is to predict the bitcoin USD price using the Long Short-Term Memory Recurrent Neural Network (LSTM-RNN) model. The LSTM-RNN model was chosen because it is better than the traditional neural network model. Measurement of the results in this study using the Root Mean Square Error (RMSE). The RMSE results obtained on the application of the LSTM-RNN model 6461.14.