Abstract:    The urban knowledge innovation network plays a fundamental supporting role in the formation of new quality productive forces. This paper utilized data from 285 cities spanning from 2011 to 2021, and within the "elements-structure-function" framework of systems theory, measures the development level of new quality productive forces. Additionally, it employed double machine learning methods to examine the specific impact of the knowledge innovation network on new quality productive forces. The main findings are as follows: (1) The knowledge innovation network positively promotes the development of new quality productive forces, particularly playing an active role in enhancing the elemental characteristics and functional orientation of productivity. (2) Further analysis reveals that factors such as intellectual property protection, elimination of information barriers, and supply of data elements play a positive moderating role in the impact of knowledge innovation networks on new quality productivity. (3) Heterogeneity testing shows that the impact of knowledge innovation networks on new quality productivity exhibits significant spatial and hierarchical heterogeneity. In the eastern regions and urban agglomerations, the promoting effect of the urban knowledge innovation network is more pronounced. Moreover, the impact of the knowledge innovation network shows a gradient characteristic, decreasing from "leading areas" to "lagging areas". This paper has enriched the research on the influencing factors of new quality productive forces and the external effects of urban knowledge innovation networks. The conclusions will provide some policy implications for cities to promote the emergence of new quality productive forces through knowledge innovation.

Key words: knowledge innovation network, new quality productive forces, moderating effect, heterogeneity test, double machine learning