查看文章

A significant amount of research is carried out on integrating historical data and sentiment analysis to predict stock prices. This study proposes DTR-SA approach to predict the next day’s high price for the randomly selected stocks from the NIFTY 50 Indian stock market index. The approach first calculates sentiment scores based on the news data collected from various news portals. The Decision Tree Regression (DTR) algorithm next combines historical stock prices with a sentiment score to predict the high price for the next day. Results indicate that sentiment analysis of news data along with historical stock prices significantly improves the prediction accuracy and hence the Mean Absolute Percentage Error (MAPE) value. The MAPE value of all stocks considered by this study is better compared to the previous historical results.