Abstract:With the continuous enhancement of innovation links between cities, the optimization and development of urban innovation networks will play an increasingly prominent role in promoting the efficiency of urban green innovation. Some studies have examined the spatial spillover effect of urban green innovation efficiency and the influence of urban innovation network characteristics, but few works of literature focus on the impact of urban network node characteristics on its green innovation efficiency, especially lack of corresponding empirical analysis. This paper believes that in the urban innovation network, each city is a node in the network. The aggregation of innovation resources and activities at nodes produces scale effects, and the flow of innovation resources and activities between nodes brings spillover effects, both of which are conducive to the improvement of urban green innovation efficiency. Therefore, the centrality of nodes, which reflects the importance of nodes in network connections, and the increase in the degree of node aggregation, which reflects the degree of clustering of different nodes in the network, are both conducive to the improvement of urban green innovation efficiency. The improvement of node centrality puts forward higher requirements for the city’s infrastructure construction, so it can promote the efficiency of urban green innovation by improving the path of infrastructure; the improvement of node aggregation degree brings more demands and opportunities for urban financial development, so it can promote the efficiency of urban green innovation by increasing the scale of innovation investment and financing. In areas with high urban density and economic development level, the innovation connection between cities is more convenient, and the improvement of node centrality has a greater role in promoting the efficiency of urban green innovation; however, in areas with low urban density and economic development level, the efficiency of innovation aggregation is relatively high, and the improvement of node aggregation degree has a greater role in promoting the improvement of urban green innovation efficiency. Taking 279 cities in China from 2010 to 2018 as a sample, the urban innovation network is constructed through the gravity model, and the node centrality and node aggregation degree are measured. The analysis found that the node centrality and node aggregation degree have a significant positive impact on the efficiency of urban green innovation; urban transportation infrastructure has a significant partial intermediary effect in the promotion of urban green innovation efficiency by node hub, and the deposit level of urban financial institutions has a significant masking effect in the promotion of urban green innovation efficiency by node agglomeration; the promotion effect of node centrality on urban green innovation efficiency weakens in turn in the eastern, central and western regions, while the promotion effect of node agglomeration increases in turn. Compared with the existing literature, this paper mainly expands and deepens this study from three aspects: one is to expand the research on the influencing factors of urban green innovation efficiency from the perspective of network node characteristics; the other is to deepen the network node characteristics from the perspective of influencing paths and regional heterogeneity; the third is to provide empirical evidence for promoting the efficiency of green innovation by improving the characteristics of network nodes. This paper reveals the important role of the evolution of urban innovation network (the improvement of node centrality and node aggregation) in improving the efficiency of green innovation, its influence path and regional heterogeneity, which is conducive to further promoting innovative development and green development by optimizing the development pattern of the urban network in the new development stage.