| Summary: | 碩士 === 國立臺東大學 === 文化資源與休閒產業學系 === 102 === With the growth of internet, the electronic word-of-mouth (eWOM) becomes an important reference influencing tourism activities. Through the characteristics of Internet reputation, this study used the Internet emotional indicators, Internet search popularity on Google Trends, and economic indicators to empirically predict
that the volume of tourism in Liuqiu is the exact acceptance of the liuqiu tourism market.
In this research, we use text mining method to collect total 3,452 discussion
records about Linqiu tourism from website’s blogs from January 2009 to December 2013. After word segmentation process, feature selection, extracting blog users’
sentiment and converting it into sentiment scores, then, the Internet search popularity
of Google Trends and five tourism-related economic indicators were taken as independen variables. The back-propagation network (BPN) of artificial neural
network was used to construct the prediction models for combinations of different
types of variables for verification and forecast accuracy evaluation. The empirical
results show: 1. Taking advantage of economic indicators , Internet search popularity on Google Trends combined with electronic word-of-mouth can really be used in forecasting the tourism market acceptances. 2. The forecast model that combines electronic word-of mouth with Internet search popularity on Google Trends is superior to artificial neural network models 3. The empirical results show that the best forecasting ability is when emotional scores of Internet reputation are moved forward for one month, indicating that the influential time point for Internet
emotion to forecast the future volume of tourism in the Linqiu tourism market is one month.
Based on the empirical results, this study provides an objective, innovative and convenient way of forecasting for the Liuqiu govermemt and manager. It can be used practically as a forecasting reference for operators when preparing plans of future operations.
|
|---|
