Our paper entitled “A graph-based approach to detect the tourist movement pattern using social media data” has been accepted by Cartography and Geographic Information Science.
This paper introduces a graph-based approach to detect the tourist movement patterns from massive and noisy social media(Twitter) data, and an object-based model is designed to represent the tourist’s spatiotemporal movement trajectory. To build the tourist graph, we first utilize the DBSCAN-based method to cluster the tourist trajectories to identify the vertices in the graph and then connect the vertices by using the tourist trajectories to generate the edges of the graph. Once the tourist graph is constructed, a set of graph-based network analysis methods is introduced to detect the tourist movement patterns.
New York City is used as the study area to demonstrate and evaluate the proposed approach. Based on the results of the case study, we reveal the tourist movement patterns by detecting the popular attractions, centric attraction, popular point-to-point routes, popular tour routes from the tourist graph. These results demonstrate that the proposed methodologies provide a feasible and effective way to build a graph-based network model for tourists from big social media data to analyse their movement patterns.
