A GIS-Based K-Mean Clustering Algorithm for Characteristic Towns in China

The unbalanced development among regions and city-towns has attracted considerable attention due to the rapid China’s urbanization process. What promote the characteristic town, a key node in China's urban system, to be the new driving force of China's urbanization development and national economic adjustment are population, living environment, rural vitality restoration and economic innovation. However, the unitary development of characteristic towns will exert an adverse impact on the long-term economic development. On the condition that previous academic work focuses less on the spatial clustering, spatial features based on the Chinese city-town system are thus introduced in this paper, and then the basic data is integrated through the Orange platform for statistical analysis. Finally, on the basis of the spatial featural visualization obtained through nearest-neighbor search, the K-mean algorithm and GIS tool are applied to perform the visualization and conduct spatial analysis of characteristic town. The results indicate that: (1) premised on the large gap in the towns’ quantity, the distance between characteristic towns and prefecture-level cities as well as major cities is quite different in diverse regions; (2) the characteristic towns can be divided into four types in space, where this four spatial clusters are gradually increasing from east to west; (3) different spatial characteristic types require various unfolding policies.