International Research Journal of Education and Technology

The stock market is influenced by complex and dynamic factors, making accurate trend prediction a challenging task. Traditional forecasting methods often fail to capture the temporal dependencies in market data. This study proposes the use of Long Short-Term Memory (LSTM) networks, a type of recurrent neural network, to improve stock market trend prediction. LSTM networks are particularly well-suited for learning from sequential data, making them ideal for forecasting stock prices based on historical data and technical indicators. The objective of this research is to develop an LSTM-based model to predict stock market trends and assist in investment decision-making. The model is trained on historical stock price data and key financial indicators. Results show that the LSTM model outperforms traditional machine learning techniques, achieving higher prediction accuracy and lower error rates. The ability of LSTM to capture long-term dependencies in the data leads to more reliable forecasts. In conclusion, LSTM networks offer a promising approach for enhancing stock market predictions, providing investors with more accurate insights for decision-making and risk management.

Key Words: Stock market prediction, LSTM, time series forecasting, machine learning, financial analysis.