StockSafad IsmailHarsha Vasudev market is a dynamic financial marketplace where individuals and organizations engage in the buying, selling, and trading of publicly traded company shares. Stock market forecasting involves estimating future changes in stock values through various analytical methods such as statistical analysis, historical data trends, and economic indicators. Traditional methods such as technical and fundamental analysis are widely used. However, sentiment analysis particularly using social media data has emerged as a promising area for enhancing prediction accuracy. This work integrates sentiment analysis from X data with advanced Machine Learning (ML) models to examine the effect of market perception on stock prices. Techniques such as Auto Regressive Integrated Moving Average (ARIMA), Moving Average (MA), and Long Short Term Memory (LSTM) networks were used to predict stock prices. Sentiment analysis were performed on X data to examine the effect of market perception on stock prices using the VADER and Text Blob lexicon sets. The result showed that adding sentiment analysis can improve the stock price prediction models accuracy. Statistical Evaluation Metrics showed that LSTM model has achieved higher correlation values, indicating superior predictive performance over the ARIMA and MA Models. Additionally, VADER outperformed Text Blob, demonstrating higher accuracy and more reliable results. The findings have significant implications for investors, financial analysts, and researchers interested in stock market prediction. They highlight the potential of sentiment analysis as a supplementary tool for understanding market behavior.