Using ANN to Analyze the Correlation Between Tourism-Related Hot Words and Tourist Numbers: A Case Study in Japan

Abstract

Google's search engine has recorded the popularity of a great number of tourism-related hot words. Prior to vacationing, many people will search the four dimensions of tourism, namely food, fashion, accommodation and transportation, on the Internet before an overseas trip. Exploring the correlation between popularity trends of tourism-related hot words and the number of tourists visiting a particular destination is a potentially valuable research area for the tourist industry. Therefore, this study counted the occurrence frequency of words related to Japanese tourism in the Google search engine and in tourism articles on electronic news websites. With these data, it calculated the Pearson correlation coefficient of the number of Taiwanese tourists visiting Japan "n" months later. Additionally, a deep learning (Artificial Neural Network) model was established, and the relationship between the popularity scores of tourism-related hot words and the interval of the number of Taiwanese tourists in Japan was examined. The research results show that the popularity of tourism-related hot words on Google is highly related to the number of Taiwanese tourists visiting Japan.