This study explores machine learning approaches for forecasting stock prices, addressing the complexities of high volatility and non-linear market behaviour. Traditional forecasting methods, while useful, lack adaptability in dynamic markets. We employ three models—Long Short-Term Memory (LSTM) networks, Random Forest, and Linear Regression—to analyze historical price patterns and incorporate sentiment analysis and economic indicators as additional predictive features. Methodology includes rigorous data preprocessing, feature engineering, and model evaluation using Mean Absolute Error (MAE) and Root Mean Square Error (RMSE) metrics. Our results indicate that LSTM achieves the highest accuracy among the models, making it a promising tool for financial forecasting. This paper contributes a comparative analysis and identifies influential factors impacting predictive performance in stock market applications.