Building a Stock Price Prediction Model using Random Forest Regression and Sentimental Analysis

This paper presents a hybrid stock price prediction model that combines Long Short-Term Memory (LSTM) networks with traditional statistical models like ARIMA, as well as real-time sentiment analysis, to improve forecasting accuracy and adaptability in volatile market conditions. Our model integrates historical stock data with external factors such as news articles and social media sentiment, providing a holistic approach to price prediction. The proposed system is evaluated using performance metrics including MAE, RMSE, and R², and shows superior results compared to standalone models. Furthermore, we provide practical implementation details using Python, TensorFlow, and real-time APIs to demonstrate the feasibility and scalability of our approach.