标题：A SYSTEMATIC REVIEW ON THE RELATIONSHIP BETWEEN STOCK MARKET PREDICTION MODEL USING SENTIMENT ANALYSIS ON TWITTER BASED ON MACHINE LEARNING METHOD AND FEATURES SELECTION
期刊名称：Journal of Theoretical and Applied Information Technology
印刷版ISSN：1992-8645
电子版ISSN：1817-3195
出版年度：2017
卷号：95
期号：24
页码：6924
出版社：Journal of Theoretical and Applied
摘要：This study is mainly a systematic review analysis which discusses studies related to the role of sentiment analysis, Twitter data, and features in predicting stock market returns. Studies show that it is not only the historical financial data of firms or stock markets that can predict the returns of the stock market, but sentiments and emotions of people can also help in predicting stock market returns. One primary source of information that is available now to everyone is Twitter, and tweets made by significant personals affect the emotions of people and which will ultimately affect their investment decisions. If the news is positive, then most probably it will affect people positively, and they will invest more in stocks of that firm. If the news is negative, the reaction of people is expected to be opposite. Besides the sentiments of people, there are features like spatial and temporal that can also affect the stock market returns. The spatial feature is a geographical division, it can either be different emotions of different people from different geographical regions, or they can be other stock markets which can affect the home stock market by any relation. Similarly, temporal effect shows the change in something over a span of time. People might have different opinions at different times, and they can behave differently according to their sentiments at that specific time. Finally, all these factors help us in predicting the future stock market returns.
