Tourism trend and network analysis utilizing big data on social media

Abstract

Background/Objectives: With evolving trends, tourism is also experiencing more diverse policies and methods of promotion. In particular, with the development and increasing popularity of social media platforms, a new trend is setting in. In line with such changes, the current study sets out to utilize big data on social media platforms to analyze trends in tourism, ways in which tourism elements mutually interact, and analyze patterns, in order to propose tourism promotion strategies and provide related basic data.Methods/Statistical analysis: Analysis on social media platforms were conducted to visually express relationship among nodes and analyze the structure and status of link in quantitative terms. NodeXL is an add-in program to Microsoft Excel; it allows the user to directly collect data from social media platforms to execute matrics, statistics, and visualization. The data was collected from Korea Tourism Organization (KTO)’s Twitter and Facebook accounts. Hashtags (#) on 3,200 posts on the Twitter account were analyzed to compute the tourism trend, and the inter-node interactions and links on the Facebook fan pages were analyzed in terms of network density and centrality to calculate the form and characteristics of social media networks.Findings: By analyzing social media pages that represent promotional efforts for Korean tourism, we were able to find the following results: On the KTO Twitter account, the higher hashtag terms were “eating tour,” and “exciting travel,” which follow the recent tourism trends. However, because of platform restrictions, the Twitter account, rather than engaging in mutual interactions with its users, only tended to deliver information, and was unable to reflect more diverse tourism trends. On Facebook, 348 nodes were actively linked 14.99 times on average, indicating a healthy level of activity. Average degrees of connection was 2.214, which is smaller than average connection distance of small societies, indicating efficient mutual interaction. There were three core user groups, with eleven individuals serving as media nodes, and six users with Eigenvector centrality.Improvements/Applications: Tourism promotion must be executed in line with diverse and latest trends in the field. Because Facebook has a higher level of mutual interaction than Twitter, the account holder can maximize the promotional effects by utilizing individuals that serve as the centrality node. That is to say that promotional strategies that take into account the characteristics of individual social media platform are required.