Crypto Price Prediction Using LSTM Models

Crypto Price Prediction Using LSTM Models

Author Name : Mrs. M.Gayathridevi, Mr. M. Praveen Eswar, Mr.G.Elamaran, Mr. S.Vignesh

ABSTRACT This paper presents a method and system for predicting cryptocurrency prices utilizing an LSTM (Long ShortTerm Memory) model. Key focus areas include data acquisition, preprocessing, and forecasting stages, where historical price data, technical indicators, and market sentiment are leveraged to enhance prediction accuracy. By structuring the data pipeline effectively, the model captures critical features in the volatile crypto market. The trained LSTM model demonstrates superior performance over traditional methods such as ARIMA and linear regression by effectively capturing long-term dependencies. This approach enables more accurate forecasts, aiding investors in making well-informed decisions. The study underscores LSTM's potential in addressing the unique challenges posed by cryptocurrency markets, which are characterized by rapid fluctuations and high uncertainty.