A Novel Popular Tourist Attraction Discovering Approach Based on Geo-Tagged Social Media Big Data

TLDR: A new method for discovering popular tourist attractions, which extracts hotspots through integrating spatial clustering and text mining approaches is proposed and the popularity distribution laws of Beijing’s tourist attractions under different temporal and weather contexts are analyzed.

Abstract: In the big data era, the social media data that contain users’ geographical locations are growing explosively. These kinds of spatiotemporal data provide a new perspective for us to observe the human movement behavior. By mining such spatiotemporal data, we can incorporate the users’ collective wisdom, build novel services and bring convenience to people. Through spatial clustering of the original user locations, both the ‘natural’ boundaries and the human activity information of the tourist attractions are generated, which facilitate performing popularity analysis of tourist attractions and extracting the travelers’ spatio-temporal patterns or travel laws. On the one hand, the potential extracted knowledge could provide decision supports to the tourism management department in both tourism planning and resource development; on the other hand, the travel preferences are able to be extracted from the clustering-generated attractions, and thus, intelligent tourism recommendation services could be developed for the tourist to promote the realization of ‘smart tourism’. Hence, this paper proposes a new method for discovering popular tourist attractions, which extracts hotspots through integrating spatial clustering and text mining approaches. We carry out tourist attraction discovery experiments based on the Flickr geotagged images within the urban area of Beijing from 2005 to 2016. The results show that compared with the traditional DBSCAN method, this novel approach can distinguish adjacent high-density areas when discovering popular tourist attractions and has better adaptability in the case of an uneven density distribution. In addition, based on the finding results of scenic hotspots, this paper analyzes the popularity distribution laws of Beijing’s tourist attractions under different temporal and weather contexts.