This document discusses a data-driven approach to stock market prediction and sentiment analysis. It proposes combining recurrent neural networks with long short-term memory (RNN-LSTM) to predict stock prices based on historical data, and using support vector machines (SVM) to analyze sentiment from news headlines and predict how it may affect stock trends. The paper reviews several related works applying machine learning techniques like RNN, LSTM, and SVM to stock prediction and sentiment analysis. It aims to improve prediction accuracy by combining both historical data analysis and sentiment analysis of news articles.