Forecasting the Fluctuations in the Price of Cryptocurrency using LSTM in Machine Learning

One of the most well-known and valuable cryptocurrencies in the present financial market is bitcoin, which attracts investors and thus creates new research opportunities for scientists. Numerous studies using various machine learning prediction methods have been conducted on predictingBitcoin prices. In order to conduct the study, important features from a dataset with a high connection to Bitcoin prices are gathered, and then random data chunks are chosen to train and test the model. The accuracy of price predictions may be lowered due to unfitting results caused by the random data that was chosen for model training. Here, a good training procedure for a prediction model is under examination. The simple Long Short-Term Memory (LSTM) model is then trained using the suggested methods to forecast the price of bitcoin. The price of bitcoin for the next five days. Sustainable outcomes for the prediction are discovered when the LSTM model is trained with an appropriate data chunk, which has been identified. Future advancements are the work's climax in this paper's conclusion. [5