A network analysis of tourist activity

Purpose: This study aims to model tourist activities in a network and explore the properties of the network. Such network enables the author to explain and quantify how tourist activities are connected in determining tourist consumption as well as the organization of destination supply. Design/methodology/approach: The author developed a network formation mechanism to create edges between nodes based on the joint probability of a pair of activities undertaken by tourists at a destination. By adjusting network sparsity, the author created an ensemble of four topologically similar networks for empirical testing. The author used tourist activity data of Hong Kong inbound tourists to test the network model. Findings: The author found a robust hub–periphery topological structure of the tourist activity network. In addition, the network is featured by high clustering, short diameter and positive correlations between four node centralities, namely, degree, closeness, betweenness and eigenvector centralities. The author also generated the k-cores of the networks to further unravel the structure of hub nodes. The author found that the k-cores are dominated by tourist activities related to shopping or sightseeing, suggesting the high complementarity of these activities. Research limitations/implications: This study provides a different lens through which tourist consumption can be understood from a macroscopic angle by examining network topology and from a microscopic angle by examining node centralities. Originality/value: To the best of the author’s knowledge, this is the first study attempting to model tourist activity and consumption in a network and explore the properties of the network. Not only has this study provided a new real-world network for network research, but it has also suggested an innovative modeling approach for tourist behavior research