Sentiment Analysis for Stock Price Prediction

Stock prices and financial markets are often sentiment-driven, which leads to research efforts to predict stock market trend using public sentiments expressed on social media such as Facebook and Twitter. In this project, we investigate the impact of sentiment expressed through StockTwits on stock price prediction. StockTwits is a relatively new microblogging website, which is becoming increasingly popular for users to share their discussions and sentiments about stocks and financial markets. Specifically, we analyze the StockTwits tweet contents and extract financial sentiment using a set of text featurization and machine learning algorithms. The correlation between the aggregated daily sentiment and daily stock price movement is then studied. Finally, the sentiment information is used in addition to the past stock time series data to improve the accuracy of stock price movement prediction. The effectiveness of the proposed work on stock price prediction is demonstrated through experiments on five companies (Apple, Amazon, General Electric, Microsoft, and Target) using nine-month StockTwits data and daily stock data.