Study on predicting movie box office based on sentiment analysis of micro-blog
Shi We · 2015
Micro-blog is an important carrier of electronic word-of-mouth,which has affected the purchase decisions of consumers and product sales of businesses.In this article,we study the problem of mining sentiment information from Sina micro-blog and investigate ways to use such information for predicting product sales performance.We analyze the affecting factors of micro-blog reviews,and establish the sentiment compute method of micro-blog based on fuzzy sentiment ontology,then put sentiment factors into the autoregressive model,present autoregressive sentiment predicting model for predicting product sales performance.Extensive experiments were conducted on a movie data set.We analyze the parameters selection,and compare our model with alternative models that do not take into account the sentiment information.Experiments confirm the effectiveness and superiority of the proposed approach.