Modelling and application of fuzzy adaptive minimum spanning tree in tourism agglomeration area division

Publication Type:
Journal Article
Knowledge-Based Systems, 2018, 143 pp. 317 - 326
Issue Date:
Full metadata record
Files in This Item:
Filename Description Size
1-s2.0-S0950705117302800-main.pdfPublished Version2.65 MB
Adobe PDF
© 2017 Elsevier B.V. Tourism agglomeration area division plays an increasingly important role in government's policy making on planning and development of tourism industry nowadays. With the development of ICT technologies, tourism “big data” such as geographic data, tourist attractions evaluation data, tourist service data and related traffic information, becomes available to access, which provides great opportunities to develop intelligent tourism decision support systems (TDSS) for related government policy making. To effectively divide tourism agglomeration areas to support tourism-planning decision, by use of the “big data” resources, a fuzzy adaptive minimum spanning tree (F-AMST) model, which integrates adaptive minimum spanning tree (AMST) method and fuzzy level evaluation method, is proposed in this study. The F-AMST model consists of three parts: F-MST generation, F-MST splitting, and clustering solution evaluation and adjustment. The proposed model is then applied to cluster 142 A-level scenic spots in mountain areas of Hebei province, China, and the optimal tourism clustering solution with seven tourism agglomeration areas is finally obtained. The result is then analysed from the following aspects: the spatial agglomeration degree of each tourism agglomeration area is analysed by use of the fuzzy dense degree based on F-AMST model; the spatial distribution type of scenic spot nodes within each agglomeration area is analysed by using the nearest neighbour index R with consideration of spatial distance factors; the scenic spot node level system of each agglomeration area is analysed by use of the node level perfection index Z considering the scenic spot node level factor; the influences of the two factors, the spatial distance and the scenic spot node level, to the agglomeration degree are then analysed by use of the correlation between R and fuzzy dense degree and that between Z and fuzzy dense degree respectively. The findings are carefully described in this paper and the results can directly support government's decision making in tourism resources planning and construction of tourism agglomeration areas so as to improve the regional tourism competitiveness.
Please use this identifier to cite or link to this item: