Content area
Full Text
Introduction
Social media marketing has received increasing attention from both academia and practitioners because it can help businesses strengthen their relationships with customers and spread information on products, services and brands (Bilgihan et al., 2014; Xiang et al., 2015). Information diffusion through Web 2.0 platforms like Twitter and Facebook have resulted in raising awareness of brands, helping customers form attitudes and even affecting their decision-making (Kwok and Yu, 2013; Mangold and Faulds, 2009). In particular, the impact of social media in the hospitality industry is significant because customers are more likely to seek personal suggestions on social media and rely on messages posted by other customers on social media (Pantelidis, 2010). Such social networks are therefore a new form of electronic word-of-mouth (eWOM) (Bruns and Burgess, 2012).
Considering the tremendous increase in social media users and consequent growth in volume of user-generated content, social media analytics has become a new method for investigating trends and patterns (Boyd and Ellison, 2007; Bruns and Burgess, 2011a). Twitter, a popular microblogging service, is used for social media analytics partly because of its popularity and because data collection is feasible (Kwak et al., 2010). Launched in 2006 (Bruns and Burgess, 2011b), Twitter had an estimated 271 million active users in 2014, with 78 per cent of them using Twitter services on their mobile devices (Twitter, 2014a). User-generated content in Twitter, or a tweet, includes diverse attributes like message text, screen name of sender, posting time, language type and so on. Given that customers provide honest opinions on products and service and that information via social media is highly valued by other customers (Burton and Khammash, 2010), social media analysis using Twitter is important to the hospitality industry.
Despite the increasing importance of using social media analytics to predict current trends and matters of common interest (Dodds et al., 2011; Weiss et al., 2010), few empirical studies analyzing tweets in the hospitality management have been conducted (Leung et al., 2013). Kwok and Yu (2013) mentioned that applying social media analytics to restaurant research is still in an early stage. Analyzing customer emotion with user-generated content (i.e. tweet messages) would allow researchers to estimate customer attitudes toward a product, service or brand. Sentiment analysis methods...