This paper proposes a novel model to enhance the prediction of stock market movements using sentiment analysis (SA) by integrating neutrosophic logic (NL) to handle uncertainty and ambiguity in social media data. The model combines SA results with historical stock market data to predict stock price fluctuations. The SA process uses NL to classify tweets, which can handle uncertain and indeterminate data more effectively than traditional SA methods. The results show that the proposed model achieves a predictive accuracy of 78.48% using the StockNet dataset, surpassing previous studies. The model employs a long short-term memory (LSTM) algorithm to forecast stock movements based on SA scores and historical data. The study highlights the importance of incorporating social media data into stock market prediction models to improve accuracy and effectiveness. The model outperforms other models, including those using traditional SA techniques and different machine learning approaches, demonstrating the effectiveness of NL in handling uncertainty and ambiguity in sentiment analysis. The results also show that the model performs well in financial metrics such as return and Sharpe ratio, indicating its potential for enhancing financial forecasting and decision-making. Future work includes exploring the impact of user profiles on SA results and incorporating data from other social media platforms.This paper proposes a novel model to enhance the prediction of stock market movements using sentiment analysis (SA) by integrating neutrosophic logic (NL) to handle uncertainty and ambiguity in social media data. The model combines SA results with historical stock market data to predict stock price fluctuations. The SA process uses NL to classify tweets, which can handle uncertain and indeterminate data more effectively than traditional SA methods. The results show that the proposed model achieves a predictive accuracy of 78.48% using the StockNet dataset, surpassing previous studies. The model employs a long short-term memory (LSTM) algorithm to forecast stock movements based on SA scores and historical data. The study highlights the importance of incorporating social media data into stock market prediction models to improve accuracy and effectiveness. The model outperforms other models, including those using traditional SA techniques and different machine learning approaches, demonstrating the effectiveness of NL in handling uncertainty and ambiguity in sentiment analysis. The results also show that the model performs well in financial metrics such as return and Sharpe ratio, indicating its potential for enhancing financial forecasting and decision-making. Future work includes exploring the impact of user profiles on SA results and incorporating data from other social media platforms.