The predictive power of online chatter
Proceedings of the eleventh ACM SIGKDD international conference on Knowledge discovery in data mining
| Hi-index | 0.00 |
This paper analyzes four bestselling books in series to find leading indicators of domestic book sales from conventional Blog information. Although Blog information is different from actual purchases, it will influence customer behaviors. They consider it would be useful for decision making of businesses and organizations. Main contributions of the paper are three fold: 1) confirms relationships between on-line information of the number of Blog reference and off-line information of book sales data; 2) observes different sales patterns by analyzing series books and identify factors and classify books into category; and 3) attempts to predict sales pattern based on sales data of three books.