Entering the stage of high-quality development,metropolitan areas,as a key link in the national new-type urbanization strategic pattern,have become an important spatial unit leading the regional development.This study takes Wuhan metropolitan area(WMA)as the research area.Based on the multi-source big data such as the mobile phone signaling and the Baidu index,this study explores the urban spatial connection pattern and influencing factors of WMA by using the factor movement model,the SOM neural network classification model,and the QAP regression analysis.The results show that:1)The spatial pattern formed by the flow of various elements in WMA is a multi-center structure,and the internal flow intensity is varied between different cities.2)SOM neural network classification model shows that the cities of WMA will be divided into four levels according to their current urban spatial connections and there will be huge gaps between different levels.3)The variances in the urban development level and the informatization level have a significantly negative impact on the intensity of factor movements among different cities,while the difference of city size and openness degree has a significantly positive impact,which jointly affects the formation of urban spatial linkage pattern in metropolitan areas.According to the above,it puts forward suggestions for optimizing of the spatial linkage pattern of the Wuhan metropolitan area.
关键词
要素流/人口流/多源数据/手机信令/SOM神经网络分类模型/空间联系格局/武汉都市圈
Key words
factor movement/population flow/multi-source big data/mobile signaling/SOM neural network classification model/spatial connection pattern/Wuhan metropolitan area
维普
万方数据