In recent years, the rapid expansion of the cryptocurrency market and its high price volatility have increased the need for accurate forecasting methods and intelligent trading systems. This study presents an intelligent trading algorithm based on Long Short-Term Memory (LSTM) neural networks for price prediction and the generation of buy and sell signals in the cryptocurrency market. The main objective of this research is to improve trading decision-making performance and reduce risks arising from unpredictable market fluctuations.The proposed model is trained using historical price data, including open, close, high, low prices, and trading volume, and is capable of extracting long-term temporal dependencies in financial time series data. In this study, the data are first preprocessed and normalized, after which the LSTM network architecture is designed and optimized. The output of the model is price trend prediction, based on which trading signals are generated.Experimental results show that the LSTM model outperforms traditional methods such as moving averages and linear regression in terms of prediction accuracy and provides better performance in generating trading signals. Furthermore, the use of this model increases investment profitability and reduces decision-making errors under volatile market conditions.Overall, the findings of this research indicate that the use of deep learning networks, particularly LSTM, can play an effective role in the development of intelligent trading systems in the cryptocurrency market and serve as a useful tool for financial analysts and investors.