There has been a rise in the development of a powerful tool that can forecast stock market trends and understand how consumers behave in the stock market, nicknamed sentiment analysis. Advanced machine learning techniques such as FinBERT, LSTM, Support Vector Machine (SVM) and Naïve Bayes are used for the impact analysis of investor sentiment on stock price movements in this research. The dataset used in the study comprises financial news, social media sentiment and company announcements to establish model performance. The experimental results show that FinBERT works better than the other classifiers proposed in sentiment classification (91.3%) and stock trend forecasting (88.7%), LSTM (82.5%), and Naïve Bayes (79.4%). Furthermore, trading strategies based on these models capitalize on sentiment from 9.2% higher return than the traditional trading strategies. Integration of real time sentiment analysis strengthens predictive accuracy and market decision making as found in the findings. It also brings out something from ESG sentiment analysis that shows that if there’s positive environmental sentiment, the stock price goes up an average of 3.8%, if there’s a negative sentiment, the price is 4.5% less. The importance of the sentiment analysis in the financial modeling is underscored by this research, and the hybrid models of combining economic indicator with sentiment trend of improved forecasting are needed.