| In recent years, the public opinion is swayed by online social, media and news platforms, such as Twitter, podcasts, and streaming news broadcasts. The public opinion can alter the outcome of various social-economic events, e.g., the volatility of the stock market. This paper presents an overview of forecasting the volatility of the indices of several companies in the U.S. stock market while considering the sentiment and features extracted from the metadata of a tweet and its author's social activity and network. The daily changes in the prices of an index in the U.S. stock market were estimated by applying several regression techniques. The results indicate a strong correlation between the approximated closing prices of the stocks in the U.S. stock market, the sentiment along with the features extracted from a tweet, and its author's activity and network. Finally, the obtained results indicate that the number of attributes did not impact the performance of the applied regression techniques. |
*** Title, author list and abstract as seen in the Camera-Ready version of the paper that was provided to Conference Committee. Small changes that may have occurred during processing by Springer may not appear in this window.