Discovering structure and influencing factors of Chinese city directed network (CCDN) from web search engine data

Abstract

Numerous studies have addressed city network structures and their influencing factors; however, few have explored the dynamics of directed city networks in the cyberspace. We leverage web search engine big data to analyze structural characteristics of the Chinese City Directed Network (CCDN) and their influencing factors from a cyberspace perspective. The findings are as follows: (1) the CCDN is the mildly disassortative network in cyberspace, the spatial distribution of city interaction follows a "high in the east, low in the west" pattern along the Hu Line; (2) The CCDN contains seven distinct communities, whose boundaries generally correspond with provincial administrative divisions, national city agglomerations, and regional cultural zones; (3) There is a difference between regional and national hub cities, with most hub cities being provincial capitals or prominent tourist cities. In particular, Chengdu surpasses Beijing in both regional and national hubness; (4) Economic growth, investment in science and technology, number of students in secondary vocational schools, and the number of subscribers of internet services are positively correlated with city hubness index, while the proportions of the primary, secondary and tertiary industries are negatively correlated. These factors display significant spatial variability. The results not only suggest novel strategies for city managers to enhance city images and attractivity in the digital era but also contributes new perspectives to the discourse on sustainable city development.