Novel Sentiment Lexica Derived from User Generating Content by Chinese Tourists in Pacific Islands

Abstract

Identification of tourists’ sentiments is relevant to the destination’s planning. Tourists generate extensive User Generated Content (UGC)—embedding their sentiments—in the form of textual data when sharing experiences on the Internet. These UGC tend to influence tourists’ decision-making, thus, representing an important data source for tourism research and planning. By obtaining data from Mafengwo and Ctrip, sentiment analysis was conducted to shed light on the sentiment tendency of Chinese tourists in seven Pacific Island Countries and Territories (PICTs). Eleven thousand two hundred four reviews were obtained between January and March 2021. The data shows that Chinese tourists’ sentiments towards the PICTs are overall positive. Yet, they pay more attention to practical issues such as transportation, visa and fees, and their sentiment orientations are influenced by tourism resources, weather, and perceived safety. Moreover, the study demonstrates that the needs of Chinese tourists in the region are influenced by their physiology, security, self-esteem, belonging, and self-actualisation needs. The study contributes to theory and practice by constructing an exclusive set of Chinese sentiment lexicons for tourism research based on data from the PICTs. This lexicon complements but also contradicts previous studies. In addition to being relevant for the studied region, it can inform similar destinations that may or may not have a relevant Chinese tourism market.

Evolution of Public Attitudes and Opinions Regarding COVID-19 Vaccination During the Vaccine Campaign in China: Year-Long Infodemiology Study of Weibo Posts

“…It has been used to measure public opinion in areas including e-commerce [34] and public health events [35]. A post was considered positive if the sentiment score was greater than 0.6 and negative if it was smaller than 0.4; otherwise, the score was considered neutral [35][36][37]. Finally, the percentage of posts in each of the 3 sentiment categories was calculated.…”