Untangling spatio‐temporal dynamics and determinants of technology transfer from a patent assignment perspective: The case of China's AI data

Abstract

This study delves into the spatio‐temporal dynamics and influencing mechanisms of technology transfer. Leveraging graph theory, we constructed a patent transfer network to understand its evolving patterns. We redefined technology transfer types, analyzed transition probabilities through Markov chain, and summarized their temporal and spatial shifts. Incorporating spatial and nonspatial methods, we explored the heterogeneity of key drivers, such as GDP and internal R&D expenditures, across regions. Our findings reveal that China's AI technology transfer network transformed from sparse to densely interconnected, with transfer types evolving from singular to diversified directions and objects. Provinces often maintain stability or transition to adjacent types, forming agglomerations of similar transfer types. GDP and internal R&D expenditures emerge as key drivers, exerting distinct impacts across regions. This study offers insights to enterprises and policymakers in developing tailored strategies for promoting technology transfer.