Predicting Solana Cryptocurrency Prices in ۲۰۲۴: A Comparative Study of LSTM and GRU Models

In this paper, we aim to predict and compare the price of cryptocurrency Solana (SOL) in ۲۰۲۴, using Long ShortTerm Memory (LSTM) and Gated Recurrent Unit (GRU) which are recurrent neural networks. Daily price data for Solana was collected from the CoinMarketCap website over a one-year period, from January ۱ to December ۳۰, ۲۰۲۴. The significance of accurate Solana price prediction is emphasized in this study, as Solana’s price has a substantial impact on the prices of meme coins within its ecosystem. Specifically, Solana’s price fluctuations influence meme coin prices with a time lag, and predicting Solana’s price can provide critical signals for forecasting trends in the meme coin market. The LSTM and GRU models were trained using time windows of ۱۰ days to capture short-term and medium-term dependencies in price trends. The performance of the models was evaluated using RMSE (Root Mean Square Error), MAE (Mean Absolute Error) and R۲ (Coefficient of Determination). The results demonstrate that the LSTM and GRU models can predict Solana’s price trends with a relatively high degree of accuracy, offering valuable insights for investors and analysts.In this paper, we aim to predict and compare the price of cryptocurrency Solana (SOL) in ۲۰۲۴, using Long ShortTerm Memory (LSTM) and Gated Recurrent Unit (GRU) which are recurrent neural networks. Daily price data for Solana was collected from the CoinMarketCap website over a one-year period, from January ۱ to December ۳۰, ۲۰۲۴. The significance of accurate Solana price prediction is emphasized in this study, as Solana’s price has a substantial impact on the prices of meme coins within its ecosystem. Specifically, Solana’s price fluctuations influence meme coin prices with a time lag, and predicting Solana’s price can provide critical signals for forecasting trends in the meme coin market. The LSTM and GRU models were trained using time windows of ۱۰ days to capture short-term and medium-term dependencies in price trends. The performance of the models was evaluated using RMSE (Root Mean Square Error), MAE (Mean Absolute Error) and R۲ (Coefficient of Determination). The results demonstrate that the LSTM and GRU models can predict Solana’s price trends with a relatively high degree of accuracy, offering valuable insights for investors and analysts.

کلیدواژه ها:

Solana ، LSTM ، GRU ، Cryptocurrency ، CoinMarketCap