Identifying driving factors of urban digital financial network—based on machine learning methods

With the continuous development of digital finance, the correlation among urban digital finance has been increasing. In this paper, we further apply machine learning methods to study the driving factors of urban digital finance networks based on the construction of urban digital finance spatial network associated with a sample of 278 cities in China. The results of network characteristics analysis show that the core-edge structure of an urban digital finance network shows the characteristics of gradual deepening and orderly distribution; the core cities show reciprocal relationships with each other, and the edge cities lack connection with each other; the core cities match the structural hole distribution and the edge cities are limited by the network capital in their development. The results of driver analysis show that year-end loan balances, science and technology expenditures and per capita gross regional product are the main drivers of urban digital financial networks.