This paper addresses the challenge of enhancing stock market movement prediction by integrating sentiment analysis (SA) with neutrosophic logic (NL). The authors propose a novel model that improves the accuracy of SA by handling uncertain and indeterminate data from social media, particularly Twitter. The model combines NL, which uses three membership functions (truth, indeterminacy, and falsity), with a long short-term memory (LSTM) deep learning algorithm to predict stock movements. The study uses the StockNet dataset, which includes tweets and historical stock market data, to evaluate the model's performance. The results show that the proposed model achieves a predictive accuracy of 78.48%, surpassing previous studies that used the same dataset. The model's effectiveness is further demonstrated through comparisons with other models, including those using different sentiment analysis techniques and machine learning algorithms. The financial performance of the model is also assessed using metrics such as cumulative return and Sharpe ratio, showing superior results compared to established models and the buy-and-hold strategy. The paper concludes by highlighting the importance of incorporating social media data and the potential of NL in improving the accuracy of stock market predictions.This paper addresses the challenge of enhancing stock market movement prediction by integrating sentiment analysis (SA) with neutrosophic logic (NL). The authors propose a novel model that improves the accuracy of SA by handling uncertain and indeterminate data from social media, particularly Twitter. The model combines NL, which uses three membership functions (truth, indeterminacy, and falsity), with a long short-term memory (LSTM) deep learning algorithm to predict stock movements. The study uses the StockNet dataset, which includes tweets and historical stock market data, to evaluate the model's performance. The results show that the proposed model achieves a predictive accuracy of 78.48%, surpassing previous studies that used the same dataset. The model's effectiveness is further demonstrated through comparisons with other models, including those using different sentiment analysis techniques and machine learning algorithms. The financial performance of the model is also assessed using metrics such as cumulative return and Sharpe ratio, showing superior results compared to established models and the buy-and-hold strategy. The paper concludes by highlighting the importance of incorporating social media data and the potential of NL in improving the accuracy of stock market predictions.