Forecasting Bitcoin Price Based on Blockchain Information Using Long-Short Term Method

Since its founding in 2008, Bitcoin (financial code: BTC) has emerged as a digital currency market cap and continues to attract investors policymakers' attention. In recent years, BTC high price volatility, substantial increase 2016, followed by significant decline 2018. Unlike stock markets, is open for 24x7 dan no closing period. It means everyone can trade it any time. However, this flexibility carries investment risk. This research attempts forecast BTC's considering the blockchain's information minimize We employ Long-Short Term Memory (LSTM), artificial Recurrent Neural Network (RNN) architecture. Its model avoid long-term problems. The data used blockchain from August 4, 2018, January 21, 2020. with 20 neurons 500 epochs smallest MSE value. Then prediction an accuracy rate of 91.07%.